Category Archives: Research Blog

The latest news on Google Research

Advances in document understanding

Posted by Sandeep Tata, Software Engineer, Google Research, Athena Team

The last few years have seen rapid progress in systems that can automatically process complex business documents and turn them into structured objects. A system that can automatically extract data from documents, e.g., receipts, insurance quotes, and financial statements, has the potential to dramatically improve the efficiency of business workflows by avoiding error-prone, manual work. Recent models, based on the Transformer architecture, have shown impressive gains in accuracy. Larger models, such as PaLM 2, are also being leveraged to further streamline these business workflows. However, the datasets used in academic literature fail to capture the challenges seen in real-world use cases. Consequently, academic benchmarks report strong model accuracy, but these same models do poorly when used for complex real-world applications.

In “VRDU: A Benchmark for Visually-rich Document Understanding”, presented at KDD 2023, we announce the release of the new Visually Rich Document Understanding (VRDU) dataset that aims to bridge this gap and help researchers better track progress on document understanding tasks. We list five requirements for a good document understanding benchmark, based on the kinds of real-world documents for which document understanding models are frequently used. Then, we describe how most datasets currently used by the research community fail to meet one or more of these requirements, while VRDU meets all of them. We are excited to announce the public release of the VRDU dataset and evaluation code under a Creative Commons license.


Benchmark requirements

First, we compared state-of-the-art model accuracy (e.g., with FormNet and LayoutLMv2) on real-world use cases to academic benchmarks (e.g., FUNSD, CORD, SROIE). We observed that state-of-the-art models did not match academic benchmark results and delivered much lower accuracy in the real world. Next, we compared typical datasets for which document understanding models are frequently used with academic benchmarks and identified five dataset requirements that allow a dataset to better capture the complexity of real-world applications:

  • Rich Schema: In practice, we see a wide variety of rich schemas for structured extraction. Entities have different data types (numeric, strings, dates, etc.) that may be required, optional, or repeated in a single document or may even be nested. Extraction tasks over simple flat schemas like (header, question, answer) do not reflect typical problems encountered in practice.
  • Layout-Rich Documents: The documents should have complex layout elements. Challenges in practical settings come from the fact that documents may contain tables, key-value pairs, switch between single-column and double-column layout, have varying font-sizes for different sections, include pictures with captions and even footnotes. Contrast this with datasets where most documents are organized in sentences, paragraphs, and chapters with section headers — the kinds of documents that are typically the focus of classic natural language processing literature on long inputs.
  • Diverse Templates: A benchmark should include different structural layouts or templates. It is trivial for a high-capacity model to extract from a particular template by memorizing the structure. However, in practice, one needs to be able to generalize to new templates/layouts, an ability that the train-test split in a benchmark should measure.
  • High-Quality OCR: Documents should have high-quality Optical Character Recognition (OCR) results. Our aim with this benchmark is to focus on the VRDU task itself and to exclude the variability brought on by the choice of OCR engine.
  • Token-Level Annotation: Documents should contain ground-truth annotations that can be mapped back to corresponding input text, so that each token can be annotated as part of the corresponding entity. This is in contrast with simply providing the text of the value to be extracted for the entity. This is key to generating clean training data where we do not have to worry about incidental matches to the given value. For instance, in some receipts, the ‘total-before-tax’ field may have the same value as the ‘total’ field if the tax amount is zero. Having token level annotations prevents us from generating training data where both instances of the matching value are marked as ground-truth for the ‘total’ field, thus producing noisy examples.


VRDU datasets and tasks

The VRDU dataset is a combination of two publicly available datasets, Registration Forms and Ad-Buy forms. These datasets provide examples that are representative of real-world use cases, and satisfy the five benchmark requirements described above.

The Ad-buy Forms dataset consists of 641 documents with political advertisement details. Each document is either an invoice or receipt signed by a TV station and a campaign group. The documents use tables, multi-columns, and key-value pairs to record the advertisement information, such as the product name, broadcast dates, total price, and release date and time.

The Registration Forms dataset consists of 1,915 documents with information about foreign agents registering with the US government. Each document records essential information about foreign agents involved in activities that require public disclosure. Contents include the name of the registrant, the address of related bureaus, the purpose of activities, and other details.

We gathered a random sample of documents from the public Federal Communications Commission (FCC) and Foreign Agents Registration Act (FARA) sites, and converted the images to text using Google Cloud's OCR. We discarded a small number of documents that were several pages long and the processing did not complete in under two minutes. This also allowed us to avoid sending very long documents for manual annotation — a task that can take over an hour for a single document. Then, we defined the schema and corresponding labeling instructions for a team of annotators experienced with document-labeling tasks.

The annotators were also provided with a few sample labeled documents that we labeled ourselves. The task required annotators to examine each document, draw a bounding box around every occurrence of an entity from the schema for each document, and associate that bounding box with the target entity. After the first round of labeling, a pool of experts were assigned to review the results. The corrected results are included in the published VRDU dataset. Please see the paper for more details on the labeling protocol and the schema for each dataset.

Existing academic benchmarks (FUNSD, CORD, SROIE, Kleister-NDA, Kleister-Charity, DeepForm) fall-short on one or more of the five requirements we identified for a good document understanding benchmark. VRDU satisfies all of them. See our paper for background on each of these datasets and a discussion on how they fail to meet one or more of the requirements.

We built four different model training sets with 10, 50, 100, and 200 samples respectively. Then, we evaluated the VRDU datasets using three tasks (described below): (1) Single Template Learning, (2) Mixed Template Learning, and (3) Unseen Template Learning. For each of these tasks, we included 300 documents in the testing set. We evaluate models using the F1 score on the testing set.

  • Single Template Learning (STL): This is the simplest scenario where the training, testing, and validation sets only contain a single template. This simple task is designed to evaluate a model’s ability to deal with a fixed template. Naturally, we expect very high F1 scores (0.90+) for this task.
  • Mixed Template Learning (MTL): This task is similar to the task that most related papers use: the training, testing, and validation sets all contain documents belonging to the same set of templates. We randomly sample documents from the datasets and construct the splits to make sure the distribution of each template is not changed during sampling.
  • Unseen Template Learning (UTL): This is the most challenging setting, where we evaluate if the model can generalize to unseen templates. For example, in the Registration Forms dataset, we train the model with two of the three templates and test the model with the remaining one. The documents in the training, testing, and validation sets are drawn from disjoint sets of templates. To our knowledge, previous benchmarks and datasets do not explicitly provide such a task designed to evaluate the model’s ability to generalize to templates not seen during training.

The objective is to be able to evaluate models on their data efficiency. In our paper, we compared two recent models using the STL, MTL, and UTL tasks and made three observations. First, unlike with other benchmarks, VRDU is challenging and shows that models have plenty of room for improvements. Second, we show that few-shot performance for even state-of-the-art models is surprisingly low with even the best models resulting in less than an F1 score of 0.60. Third, we show that models struggle to deal with structured repeated fields and perform particularly poorly on them.


Conclusion

We release the new Visually Rich Document Understanding (VRDU) dataset that helps researchers better track progress on document understanding tasks. We describe why VRDU better reflects practical challenges in this domain. We also present experiments showing that VRDU tasks are challenging, and recent models have substantial headroom for improvements compared to the datasets typically used in the literature with F1 scores of 0.90+ being typical. We hope the release of the VRDU dataset and evaluation code helps research teams advance the state of the art in document understanding.


Acknowledgements

Many thanks to Zilong Wang, Yichao Zhou, Wei Wei, and Chen-Yu Lee, who co-authored the paper along with Sandeep Tata. Thanks to Marc Najork, Riham Mansour and numerous partners across Google Research and the Cloud AI team for providing valuable insights. Thanks to John Guilyard for creating the animations in this post.

Source: Google AI Blog


AdaTape: Foundation model with adaptive computation and dynamic read-and-write

Posted by Fuzhao Xue, Research Intern, and Mostafa Dehghani, Research Scientist, Google

Adaptive computation refers to the ability of a machine learning system to adjust its behavior in response to changes in the environment. While conventional neural networks have a fixed function and computation capacity, i.e., they spend the same number of FLOPs for processing different inputs, a model with adaptive and dynamic computation modulates the computational budget it dedicates to processing each input, depending on the complexity of the input.

Adaptive computation in neural networks is appealing for two key reasons. First, the mechanism that introduces adaptivity provides an inductive bias that can play a key role in solving some challenging tasks. For instance, enabling different numbers of computational steps for different inputs can be crucial in solving arithmetic problems that require modeling hierarchies of different depths. Second, it gives practitioners the ability to tune the cost of inference through greater flexibility offered by dynamic computation, as these models can be adjusted to spend more FLOPs processing a new input.

Neural networks can be made adaptive by using different functions or computation budgets for various inputs. A deep neural network can be thought of as a function that outputs a result based on both the input and its parameters. To implement adaptive function types, a subset of parameters are selectively activated based on the input, a process referred to as conditional computation. Adaptivity based on the function type has been explored in studies on mixture-of-experts, where the sparsely activated parameters for each input sample are determined through routing.

Another area of research in adaptive computation involves dynamic computation budgets. Unlike in standard neural networks, such as T5, GPT-3, PaLM, and ViT, whose computation budget is fixed for different samples, recent research has demonstrated that adaptive computation budgets can improve performance on tasks where transformers fall short. Many of these works achieve adaptivity by using dynamic depth to allocate the computation budget. For example, the Adaptive Computation Time (ACT) algorithm was proposed to provide an adaptive computational budget for recurrent neural networks. The Universal Transformer extends the ACT algorithm to transformers by making the computation budget dependent on the number of transformer layers used for each input example or token. Recent studies, like PonderNet, follow a similar approach while improving the dynamic halting mechanisms.

In the paper “Adaptive Computation with Elastic Input Sequence”, we introduce a new model that utilizes adaptive computation, called AdaTape. This model is a Transformer-based architecture that uses a dynamic set of tokens to create elastic input sequences, providing a unique perspective on adaptivity in comparison to previous works. AdaTape uses an adaptive tape reading mechanism to determine a varying number of tape tokens that are added to each input based on input’s complexity. AdaTape is very simple to implement, provides an effective knob to increase the accuracy when needed, but is also much more efficient compared to other adaptive baselines because it directly injects adaptivity into the input sequence instead of the model depth. Finally, Adatape offers better performance on standard tasks, like image classification, as well as algorithmic tasks, while maintaining a favorable quality and cost tradeoff.


Adaptive computation transformer with elastic input sequence

AdaTape uses both the adaptive function types and a dynamic computation budget. Specifically, for a batch of input sequences after tokenization (e.g., a linear projection of non-overlapping patches from an image in the vision transformer), AdaTape uses a vector representing each input to dynamically select a variable-sized sequence of tape tokens.

AdaTape uses a bank of tokens, called a “tape bank”, to store all the candidate tape tokens that interact with the model through the adaptive tape reading mechanism. We explore two different methods for creating the tape bank: an input-driven bank and a learnable bank.

The general idea of the input-driven bank is to extract a bank of tokens from the input while employing a different approach than the original model tokenizer for mapping the raw input to a sequence of input tokens. This enables dynamic, on-demand access to information from the input that is obtained using a different point of view, e.g., a different image resolution or a different level of abstraction.

In some cases, tokenization in a different level of abstraction is not possible, thus an input-driven tape bank is not feasible, such as when it's difficult to further split each node in a graph transformer. To address this issue, AdaTape offers a more general approach for generating the tape bank by using a set of trainable vectors as tape tokens. This approach is referred to as the learnable bank and can be viewed as an embedding layer where the model can dynamically retrieve tokens based on the complexity of the input example. The learnable bank enables AdaTape to generate a more flexible tape bank, providing it with the ability to dynamically adjust its computation budget based on the complexity of each input example, e.g., more complex examples retrieve more tokens from the bank, which let the model not only use the knowledge stored in the bank, but also spend more FLOPs processing it, since the input is now larger.

Finally, the selected tape tokens are appended to the original input and fed to the following transformer layers. For each transformer layer, the same multi-head attention is used across all input and tape tokens. However, two different feed-forward networks (FFN) are used: one for all tokens from the original input and the other for all tape tokens. We observed slightly better quality by using separate feed-forward networks for input and tape tokens.

An overview of AdaTape. For different samples, we pick a variable number of different tokens from the tape bank. The tape bank can be driven from input, e.g., by extracting some extra fine-grained information or it can be a set of trainable vectors. Adaptive tape reading is used to recursively select different sequences of tape tokens, with variable lengths, for different inputs. These tokens are then simply appended to inputs and fed to the transformer encoder.

AdaTape provides helpful inductive bias

We evaluate AdaTape on parity, a very challenging task for the standard Transformer, to study the effect of inductive biases in AdaTape. With the parity task, given a sequence 1s, 0s, and -1s, the model has to predict the evenness or oddness of the number of 1s in the sequence. Parity is the simplest non-counter-free or periodic regular language, but perhaps surprisingly, the task is unsolvable by the standard Transformer.

Evaluation on the parity task. The standard Transformer and Universal Transformer were unable to perform this task, both showing performance at the level of a random guessing baseline.

Despite being evaluated on short, simple sequences, both the standard Transformer and Universal Transformers were unable to perform the parity task as they are unable to maintain a counter within the model. However, AdaTape outperforms all baselines, as it incorporates a lightweight recurrence within its input selection mechanism, providing an inductive bias that enables the implicit maintenance of a counter, which is not possible in standard Transformers.


Evaluation on image classification

We also evaluate AdaTape on the image classification task. To do so, we trained AdaTape on ImageNet-1K from scratch. The figure below shows the accuracy of AdaTape and the baseline methods, including A-ViT, and the Universal Transformer ViT (UViT and U2T) versus their speed (measured as number of images, processed by each code, per second). In terms of quality and cost tradeoff, AdaTape performs much better than the alternative adaptive transformer baselines. In terms of efficiency, larger AdaTape models (in terms of parameter count) are faster than smaller baselines. Such results are consistent with the finding from previous work that shows that the adaptive model depth architectures are not well suited for many accelerators, like the TPU.

We evaluate AdaTape by training on ImageNet from scratch. For A-ViT, we not only report their results from the paper but also re-implement A-ViT by training from scratch, i.e., A-ViT(Ours).

A study of AdaTape’s behavior

In addition to its performance on the parity task and ImageNet-1K, we also evaluated the token selection behavior of AdaTape with an input-driven bank on the JFT-300M validation set. To better understand the model's behavior, we visualized the token selection results on the input-driven bank as heatmaps, where lighter colors mean that position is more frequently selected. The heatmaps reveal that AdaTape more frequently picks the central patches. This aligns with our prior knowledge, as central patches are typically more informative — especially in the context of datasets with natural images, where the main object is in the middle of the image. This result highlights the intelligence of AdaTape, as it can effectively identify and prioritize more informative patches to improve its performance.

We visualize the tape token selection heatmap of AdaTape-B/32 (left) and AdaTape-B/16 (right). The hotter / lighter color means the patch at this position is more frequently selected.

Conclusion

AdaTape is characterized by elastic sequence lengths generated by the adaptive tape reading mechanism. This also introduces a new inductive bias that enables AdaTape to have the potential to solve tasks that are challenging for both standard transformers and existing adaptive transformers. By conducting comprehensive experiments on image recognition benchmarks, we demonstrate that AdaTape outperforms standard transformers and adaptive architecture transformers when computation is held constant.


Acknowledgments

One of the authors of this post, Mostafa Dehghani, is now at Google DeepMind.

Source: Google AI Blog


Multimodal medical AI

Posted by Greg Corrado, Head of Health AI, Google Research, and Yossi Matias, VP, Engineering and Research, Google Research

Medicine is an inherently multimodal discipline. When providing care, clinicians routinely interpret data from a wide range of modalities including medical images, clinical notes, lab tests, electronic health records, genomics, and more. Over the last decade or so, AI systems have achieved expert-level performance on specific tasks within specific modalities — some AI systems processing CT scans, while others analyzing high magnification pathology slides, and still others hunting for rare genetic variations. The inputs to these systems tend to be complex data such as images, and they typically provide structured outputs, whether in the form of discrete grades or dense image segmentation masks. In parallel, the capacities and capabilities of large language models (LLMs) have become so advanced that they have demonstrated comprehension and expertise in medical knowledge by both interpreting and responding in plain language. But how do we bring these capabilities together to build medical AI systems that can leverage information from all these sources?

In today’s blog post, we outline a spectrum of approaches to bringing multimodal capabilities to LLMs and share some exciting results on the tractability of building multimodal medical LLMs, as described in three recent research papers. The papers, in turn, outline how to introduce de novo modalities to an LLM, how to graft a state-of-the-art medical imaging foundation model onto a conversational LLM, and first steps towards building a truly generalist multimodal medical AI system. If successfully matured, multimodal medical LLMs might serve as the basis of new assistive technologies spanning professional medicine, medical research, and consumer applications. As with our prior work, we emphasize the need for careful evaluation of these technologies in collaboration with the medical community and healthcare ecosystem.


A spectrum of approaches

Several methods for building multimodal LLMs have been proposed in recent months [1, 2, 3], and no doubt new methods will continue to emerge for some time. For the purpose of understanding the opportunities to bring new modalities to medical AI systems, we’ll consider three broadly defined approaches: tool use, model grafting, and generalist systems.

The spectrum of approaches to building multimodal LLMs range from having the LLM use existing tools or models, to leveraging domain-specific components with an adapter, to joint modeling of a multimodal model.

Tool use

In the tool use approach, one central medical LLM outsources analysis of data in various modalities to a set of software subsystems independently optimized for those tasks: the tools. The common mnemonic example of tool use is teaching an LLM to use a calculator rather than do arithmetic on its own. In the medical space, a medical LLM faced with a chest X-ray could forward that image to a radiology AI system and integrate that response. This could be accomplished via application programming interfaces (APIs) offered by subsystems, or more fancifully, two medical AI systems with different specializations engaging in a conversation.

This approach has some important benefits. It allows maximum flexibility and independence between subsystems, enabling health systems to mix and match products between tech providers based on validated performance characteristics of subsystems. Moreover, human-readable communication channels between subsystems maximize auditability and debuggability. That said, getting the communication right between independent subsystems can be tricky, narrowing the information transfer, or exposing a risk of miscommunication and information loss.


Model grafting

A more integrated approach would be to take a neural network specialized for each relevant domain, and adapt it to plug directly into the LLM — grafting the visual model onto the core reasoning agent. In contrast to tool use where the specific tool(s) used are determined by the LLM, in model grafting the researchers may choose to use, refine, or develop specific models during development. In two recent papers from Google Research, we show that this is in fact feasible. Neural LLMs typically process text by first mapping words into a vector embedding space. Both papers build on the idea of mapping data from a new modality into the input word embedding space already familiar to the LLM. The first paper, “Multimodal LLMs for health grounded in individual-specific data”, shows that asthma risk prediction in the UK Biobank can be improved if we first train a neural network classifier to interpret spirograms (a modality used to assess breathing ability) and then adapt the output of that network to serve as input into the LLM.

The second paper, “ELIXR: Towards a general purpose X-ray artificial intelligence system through alignment of large language models and radiology vision encoders”, takes this same tack, but applies it to full-scale image encoder models in radiology. Starting with a foundation model for understanding chest X-rays, already shown to be a good basis for building a variety of classifiers in this modality, this paper describes training a lightweight medical information adapter that re-expresses the top layer output of the foundation model as a series of tokens in the LLM’s input embeddings space. Despite fine-tuning neither the visual encoder nor the language model, the resulting system displays capabilities it wasn’t trained for, including semantic search and visual question answering.

Our approach to grafting a model works by training a medical information adapter that maps the output of an existing or refined image encoder into an LLM-understandable form.

Model grafting has a number of advantages. It uses relatively modest computational resources to train the adapter layers but allows the LLM to build on existing highly-optimized and validated models in each data domain. The modularization of the problem into encoder, adapter, and LLM components can also facilitate testing and debugging of individual software components when developing and deploying such a system. The corresponding disadvantages are that the communication between the specialist encoder and the LLM is no longer human readable (being a series of high dimensional vectors), and the grafting procedure requires building a new adapter for not just every domain-specific encoder, but also every revision of each of those encoders.


Generalist systems

The most radical approach to multimodal medical AI is to build one integrated, fully generalist system natively capable of absorbing information from all sources. In our third paper in this area, “Towards Generalist Biomedical AI”, rather than having separate encoders and adapters for each data modality, we build on PaLM-E, a recently published multimodal model that is itself a combination of a single LLM (PaLM) and a single vision encoder (ViT). In this set up, text and tabular data modalities are covered by the LLM text encoder, but now all other data are treated as an image and fed to the vision encoder.

Med-PaLM M is a large multimodal generative model that flexibly encodes and interprets biomedical data including clinical language, imaging, and genomics with the same model weights.

We specialize PaLM-E to the medical domain by fine-tuning the complete set of model parameters on medical datasets described in the paper. The resulting generalist medical AI system is a multimodal version of Med-PaLM that we call Med-PaLM M. The flexible multimodal sequence-to-sequence architecture allows us to interleave various types of multimodal biomedical information in a single interaction. To the best of our knowledge, it is the first demonstration of a single unified model that can interpret multimodal biomedical data and handle a diverse range of tasks using the same set of model weights across all tasks (detailed evaluations in the paper).

This generalist-system approach to multimodality is both the most ambitious and simultaneously most elegant of the approaches we describe. In principle, this direct approach maximizes flexibility and information transfer between modalities. With no APIs to maintain compatibility across and no proliferation of adapter layers, the generalist approach has arguably the simplest design. But that same elegance is also the source of some of its disadvantages. Computational costs are often higher, and with a unitary vision encoder serving a wide range of modalities, domain specialization or system debuggability could suffer.


The reality of multimodal medical AI

To make the most of AI in medicine, we’ll need to combine the strength of expert systems trained with predictive AI with the flexibility made possible through generative AI. Which approach (or combination of approaches) will be most useful in the field depends on a multitude of as-yet unassessed factors. Is the flexibility and simplicity of a generalist model more valuable than the modularity of model grafting or tool use? Which approach gives the highest quality results for a specific real-world use case? Is the preferred approach different for supporting medical research or medical education vs. augmenting medical practice? Answering these questions will require ongoing rigorous empirical research and continued direct collaboration with healthcare providers, medical institutions, government entities, and healthcare industry partners broadly. We look forward to finding the answers together.

Source: Google AI Blog


In search of a generalizable method for source-free domain adaptation

Posted by Eleni Triantafillou, Research Scientist, and Malik Boudiaf, Student Researcher, Google

Deep learning has recently made tremendous progress in a wide range of problems and applications, but models often fail unpredictably when deployed in unseen domains or distributions. Source-free domain adaptation (SFDA) is an area of research that aims to design methods for adapting a pre-trained model (trained on a “source domain”) to a new “target domain”, using only unlabeled data from the latter.

Designing adaptation methods for deep models is an important area of research. While the increasing scale of models and training datasets has been a key ingredient to their success, a negative consequence of this trend is that training such models is increasingly computationally expensive, out of reach for certain practitioners and also harmful for the environment. One avenue to mitigate this issue is through designing techniques that can leverage and reuse already trained models for tackling new tasks or generalizing to new domains. Indeed, adapting models to new tasks is widely studied under the umbrella of transfer learning.

SFDA is a particularly practical area of this research because several real-world applications where adaptation is desired suffer from the unavailability of labeled examples from the target domain. In fact, SFDA is enjoying increasing attention [1, 2, 3, 4]. However, albeit motivated by ambitious goals, most SFDA research is grounded in a very narrow framework, considering simple distribution shifts in image classification tasks.

In a significant departure from that trend, we turn our attention to the field of bioacoustics, where naturally-occurring distribution shifts are ubiquitous, often characterized by insufficient target labeled data, and represent an obstacle for practitioners. Studying SFDA in this application can, therefore, not only inform the academic community about the generalizability of existing methods and identify open research directions, but can also directly benefit practitioners in the field and aid in addressing one of the biggest challenges of our century: biodiversity preservation.

In this post, we announce “In Search for a Generalizable Method for Source-Free Domain Adaptation”, appearing at ICML 2023. We show that state-of-the-art SFDA methods can underperform or even collapse when confronted with realistic distribution shifts in bioacoustics. Furthermore, existing methods perform differently relative to each other than observed in vision benchmarks, and surprisingly, sometimes perform worse than no adaptation at all. We also propose NOTELA, a new simple method that outperforms existing methods on these shifts while exhibiting strong performance on a range of vision datasets. Overall, we conclude that evaluating SFDA methods (only) on the commonly-used datasets and distribution shifts leaves us with a myopic view of their relative performance and generalizability. To live up to their promise, SFDA methods need to be tested on a wider range of distribution shifts, and we advocate for considering naturally-occurring ones that can benefit high-impact applications.


Distribution shifts in bioacoustics

Naturally-occurring distribution shifts are ubiquitous in bioacoustics. The largest labeled dataset for bird songs is Xeno-Canto (XC), a collection of user-contributed recordings of wild birds from across the world. Recordings in XC are “focal”: they target an individual captured in natural conditions, where the song of the identified bird is at the foreground. For continuous monitoring and tracking purposes, though, practitioners are often more interested in identifying birds in passive recordings (“soundscapes”), obtained through omnidirectional microphones. This is a well-documented problem that recent work shows is very challenging. Inspired by this realistic application, we study SFDA in bioacoustics using a bird species classifier that was pre-trained on XC as the source model, and several “soundscapes” coming from different geographical locations — Sierra Nevada (S. Nevada); Powdermill Nature Reserve, Pennsylvania, USA; Hawai’i; Caples Watershed, California, USA; Sapsucker Woods, New York, USA (SSW); and Colombia — as our target domains.

This shift from the focalized to the passive domain is substantial: the recordings in the latter often feature much lower signal-to-noise ratio, several birds vocalizing at once, and significant distractors and environmental noise, like rain or wind. In addition, different soundscapes originate from different geographical locations, inducing extreme label shifts since a very small portion of the species in XC will appear in a given location. Moreover, as is common in real-world data, both the source and target domains are significantly class imbalanced, because some species are significantly more common than others. In addition, we consider a multi-label classification problem since there may be several birds identified within each recording, a significant departure from the standard single-label image classification scenario where SFDA is typically studied.

Illustration of the "focal → soundscapes" shift. In the focalized domain, recordings are typically composed of a single bird vocalization in the foreground, captured with high signal-to-noise ratio (SNR), though there may be other birds vocalizing in the background. On the other hand, soundscapes contain recordings from omnidirectional microphones and can be composed of multiple birds vocalizing simultaneously, as well as environmental noises from insects, rain, cars, planes, etc.

Audio files                 Focal domain
      		           Soundscape domain1
     
Spectogram images                 
Illustration of the distribution shift from the focal domain (left) to the soundscape domain (right), in terms of the audio files (top) and spectrogram images (bottom) of a representative recording from each dataset. Note that in the second audio clip, the bird song is very faint; a common property in soundscape recordings where bird calls aren’t at the “foreground”. Credits: Left: XC recording by Sue Riffe (CC-BY-NC license). Right: Excerpt from a recording made available by Kahl, Charif, & Klinck. (2022) "A collection of fully-annotated soundscape recordings from the Northeastern United States" [link] from the SSW soundscape dataset (CC-BY license).

State-of-the-art SFDA models perform poorly on bioacoustics shifts

As a starting point, we benchmark six state-of-the-art SFDA methods on our bioacoustics benchmark, and compare them to the non-adapted baseline (the source model). Our findings are surprising: without exception, existing methods are unable to consistently outperform the source model on all target domains. In fact, they often underperform it significantly.

As an example, Tent, a recent method, aims to make models produce confident predictions for each example by reducing the uncertainty of the model's output probabilities. While Tent performs well in various tasks, it doesn't work effectively for our bioacoustics task. In the single-label scenario, minimizing entropy forces the model to choose a single class for each example confidently. However, in our multi-label scenario, there's no such constraint that any class should be selected as being present. Combined with significant distribution shifts, this can cause the model to collapse, leading to zero probabilities for all classes. Other benchmarked methods like SHOT, AdaBN, Tent, NRC, DUST and Pseudo-Labelling, which are strong baselines for standard SFDA benchmarks, also struggle with this bioacoustics task.

Evolution of the test mean average precision (mAP), a standard metric for multilabel classification, throughout the adaptation procedure on the six soundscape datasets. We benchmark our proposed NOTELA and Dropout Student (see below), as well as SHOT, AdaBN, Tent, NRC, DUST and Pseudo-Labelling. Aside from NOTELA, all other methods fail to consistently improve the source model.

Introducing NOisy student TEacher with Laplacian Adjustment (NOTELA)

Nonetheless, a surprisingly positive result stands out: the less celebrated Noisy Student principle appears promising. This unsupervised approach encourages the model to reconstruct its own predictions on some target dataset, but under the application of random noise. While noise may be introduced through various channels, we strive for simplicity and use model dropout as the only noise source: we therefore refer to this approach as Dropout Student (DS). In a nutshell, it encourages the model to limit the influence of individual neurons (or filters) when making predictions on a specific target dataset.

DS, while effective, faces a model collapse issue on various target domains. We hypothesize this happens because the source model initially lacks confidence in those target domains. We propose improving DS stability by using the feature space directly as an auxiliary source of truth. NOTELA does this by encouraging similar pseudo-labels for nearby points in the feature space, inspired by NRC's method and Laplacian regularization. This simple approach is visualized below, and consistently and significantly outperforms the source model in both audio and visual tasks.

NOTELA in action. The audio recordings are forwarded through the full model to obtain a first set of predictions, which are then refined through Laplacian regularization, a form of post-processing based on clustering nearby points. Finally, the refined predictions are used as targets for the noisy model to reconstruct.

Conclusion

The standard artificial image classification benchmarks have inadvertently limited our understanding of the true generalizability and robustness of SFDA methods. We advocate for broadening the scope and adopt a new assessment framework that incorporates naturally-occurring distribution shifts from bioacoustics. We also hope that NOTELA serves as a robust baseline to facilitate research in that direction. NOTELA’s strong performance perhaps points to two factors that can lead to developing more generalizable models: first, developing methods with an eye towards harder problems and second, favoring simple modeling principles. However, there is still future work to be done to pinpoint and comprehend existing methods’ failure modes on harder problems. We believe that our research represents a significant step in this direction, serving as a foundation for designing SFDA methods with greater generalizability.


Acknowledgements

One of the authors of this post, Eleni Triantafillou, is now at Google DeepMind. We are posting this blog post on behalf of the authors of the NOTELA paper: Malik Boudiaf, Tom Denton, Bart van Merriënboer, Vincent Dumoulin*, Eleni Triantafillou* (where * denotes equal contribution). We thank our co-authors for the hard work on this paper and the rest of the Perch team for their support and feedback.

1Note that in this audio clip, the bird song is very faint; a common property in soundscape recordings where bird calls aren’t at the “foreground”. ↩

Source: Google AI Blog


Google at ICML 2023

Posted by Cat Armato, Program Manager, Google

Groups across Google actively pursue research in the field of machine learning (ML), ranging from theory and application. We build ML systems to solve deep scientific and engineering challenges in areas of language, music, visual processing, algorithm development, and more. We aim to build a more collaborative ecosystem with the broader ML research community through open-sourcing tools and datasets, publishing our work, and actively participating in conferences.

Google is proud to be a Diamond Sponsor of the 40th International Conference on Machine Learning (ICML 2023), a premier annual conference, which is being held this week in Honolulu, Hawaii. As a leader in ML research, Google has a strong presence at this year’s conference with over 120 accepted papers and active involvement in a number of workshops and tutorials. Google is also proud to be a Platinum Sponsor for both the LatinX in AI and Women in Machine Learning workshops. We look forward to sharing some of our extensive ML research and expanding our partnership with the broader ML research community.

Registered for ICML 2023? We hope you’ll visit the Google booth to learn more about the exciting work, creativity, and fun that goes into solving a portion of the field’s most interesting challenges. Visit the @GoogleAI Twitter account to find out about Google booth activities (e.g., demos and Q&A sessions). See Google DeepMind’s blog to learn about their technical participation at ICML 2023.

Take a look below to learn more about the Google research being presented at ICML 2023 (Google affiliations in bold).



Board and Organizing Committee

Board Members include: Corinna Cortes, Hugo Larochelle
Tutorial Chairs include: Hanie Sedghi



Google Research booth activities

Presenters: Bryan Perozzi, Anton Tsitsulin, Brandon Mayer
Title: Unsupervised Graph Embedding @ Google (paper, EXPO workshop)
Tuesday, July 25th at 10:30 AM HST

Presenters: Zheng Xu
Title: Federated Learning of Gboard Language Models with Differential Privacy (paper 1, paper 2, blog post)
Tuesday, July 25th at 3:30 PM HST

Presenters: Thomas Kipf
Title: Self-supervised scene understanding (paper 1, paper 2)
Wednesday, July 26th at 10:30 AM HST

Presenters: Johannes von Oswald, Max Vladymyrov
Title: Transformers learn in-context by gradient descent (paper)
Wednesday, July 26th at 3:30 PM HST



Accepted papers

Scaling Vision Transformers to 22 Billion Parameters (see blog post)
Mostafa Dehghani, Josip Djolonga, Basil Mustafa, Piotr Padlewski, Jonathan Heek, Justin Gilmer, Andreas Steiner, Mathilde Caron, Robert Geirhos, Ibrahim Alabdulmohsin, Rodolphe Jenatton, Lucas Beyer, Michael Tschannen, Anurag Arnab, Xiao Wang, Carlos Riquelme, Matthias Minderer, Joan Puigcerver, Utku Evci, Manoj Kumar, Sjoerd van Steenkiste, Gamaleldin F. Elsayed, Aravindh Mahendran, Fisher Yu, Avital Oliver, Fantine Huot, Jasmijn Bastings, Mark Patrick Collier, Alexey Gritsenko, Vighnesh Birodkar, Cristina Vasconcelos, Yi Tay, Thomas Mensink, Alexander Kolesnikov, Filip Pavetić, Dustin Tran, Thomas Kipf, Mario Lučić, Xiaohua Zhai, Daniel Keysers, Jeremiah Harmsen, Neil Houlsby

Fast Inference from Transformers via Speculative Decoding
Yaniv Leviathan, Matan Kalman, Yossi Matias

Best of Both Worlds Policy Optimization
Christoph Dann, Chen-Yu Wei, Julian Zimmert

Inflow, Outflow, and Reciprocity in Machine Learning
Mukund Sundararajan, Walid Krichene

Transformers Learn In-Context by Gradient Descent
Johannes von Oswald, Eyvind Niklasson, Ettore Randazzo, João Sacramento, Alexander Mordvintsev, Andrey Zhmoginov, Max Vladymyrov

Arithmetic Sampling: Parallel Diverse Decoding for Large Language Models
Luke Vilnis, Yury Zemlyanskiy, Patrick Murray*, Alexandre Passos*, Sumit Sanghai

Differentially Private Hierarchical Clustering with Provable Approximation Guarantees (see blog post)
Jacob Imola*, Alessandro Epasto, Mohammad Mahdian, Vincent Cohen-Addad, Vahab Mirrokni

Multi-Epoch Matrix Factorization Mechanisms for Private Machine Learning
Christopher A. Choquette-Choo, H. Brendan McMahan, Keith Rush, Abhradeep Thakurta

Random Classification Noise Does Not Defeat All Convex Potential Boosters Irrespective of Model Choice
Yishay Mansour, Richard Nock, Robert Williamson

Simplex Random Features
Isaac Reid, Krzysztof Choromanski, Valerii Likhosherstov, Adrian Weller

Pix2Struct: Screenshot Parsing as Pretraining for Visual Language Understanding
Kenton Lee, Mandar Joshi, Iulia Turc, Hexiang Hu, Fangyu Liu, Julian Eisenschlos, Urvashi Khandelwal, Peter Shaw, Ming-Wei Chang, Kristina Toutanova

Mu2SLAM: Multitask, Multilingual Speech and Language Models
Yong Cheng, Yu Zhang, Melvin Johnson, Wolfgang Macherey, Ankur Bapna

Robust Budget Pacing with a Single Sample
Santiago Balseiro, Rachitesh Kumar*, Vahab Mirrokni, Balasubramanian Sivan, Di Wang

A Statistical Perspective on Retrieval-Based Models
Soumya Basu, Ankit Singh Rawat, Manzil Zaheer

Approximately Optimal Core Shapes for Tensor Decompositions
Mehrdad Ghadiri, Matthew Fahrbach, Gang Fu, Vahab Mirrokni

Efficient List-Decodable Regression Using Batches
Abhimanyu Das, Ayush Jain*, Weihao Kong, Rajat Sen

Efficient Training of Language Models Using Few-Shot Learning
Sashank J. Reddi, Sobhan Miryoosefi, Stefani Karp, Shankar Krishnan, Satyen Kale, Seungyeon Kim, Sanjiv Kumar

Fully Dynamic Submodular Maximization Over Matroids
Paul Duetting, Federico Fusco, Silvio Lattanzi, Ashkan Norouzi-Fard, Morteza Zadimoghaddam

GFlowNet-EM for Learning Compositional Latent Variable Models
Edward J Hu, Nikolay Malkin, Moksh Jain, Katie Everett, Alexandros Graikos, Yoshua Bengio

Improved Online Learning Algorithms for CTR Prediction in Ad Auctions
Zhe Feng, Christopher Liaw, Zixin Zhou

Large Language Models Struggle to Learn Long-Tail Knowledge
Nikhil Kandpal, Haikang Deng, Adam Roberts, Eric Wallace, Colin Raffel

Multi-channel Autobidding with Budget and ROI Constraints
Yuan Deng, Negin Golrezaei, Patrick Jaillet, Jason Cheuk Nam Liang, Vahab Mirrokni

Multi-layer Neural Networks as Trainable Ladders of Hilbert Spaces
Zhengdao Chen

On User-Level Private Convex Optimization
Badih Ghazi, Pritish Kamath, Ravi Kumar, Raghu Meka, Pasin Manurangsi, Chiyuan Zhang

PAC Generalization via Invariant Representations
Advait U Parulekar, Karthikeyan Shanmugam, Sanjay Shakkottai

Regularization and Variance-Weighted Regression Achieves Minimax Optimality in Linear MDPs: Theory and Practice
Toshinori Kitamura, Tadashi Kozuno, Yunhao Tang, Nino Vieillard, Michal Valko, Wenhao Yang, Jincheng Mei, Pierre Menard, Mohammad Gheshlaghi Azar, Remi Munos, Olivier Pietquin, Matthieu Geist,Csaba Szepesvari, Wataru Kumagai, Yutaka Matsuo

Speeding Up Bellman Ford via Minimum Violation Permutations
Silvio Lattanzi, Ola Svensson, Sergei Vassilvitskii

Statistical Indistinguishability of Learning Algorithms
Alkis Kalavasis, Amin Karbasi, Shay Moran, Grigoris Velegkas

Test-Time Adaptation with Slot-Centric Models
Mihir Prabhudesai, Anirudh Goyal, Sujoy Paul, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Gaurav Aggarwal, Thomas Kipf, Deepak Pathak, Katerina Fragkiadaki>

Algorithms for Bounding Contribution for Histogram Estimation Under User-Level Privacy
Yuhan Liu*, Ananda Theertha Suresh, Wennan Zhu, Peter Kairouz, Marco Gruteser

Bandit Online Linear Optimization with Hints and Queries
Aditya Bhaskara, Ashok Cutkosky, Ravi Kumar, Manish Purohit

CLUTR: Curriculum Learning via Unsupervised Task Representation Learning
Abdus Salam Azad, Izzeddin Gur, Jasper Emhoff, Nathaniel Alexis, Aleksandra Faust, Pieter Abbeel, Ion Stoica

CSP: Self-Supervised Contrastive Spatial Pre-training for Geospatial-Visual Representations
Gengchen Mai, Ni Lao, Yutong He, Jiaming Song, Stefano Ermon

Ewald-Based Long-Range Message Passing for Molecular Graphs
Arthur Kosmala, Johannes Gasteiger, Nicholas Gao, Stephan Günnemann

Fast (1+ε)-Approximation Algorithms for Binary Matrix Factorization
Ameya Velingker, Maximilian Vötsch, David Woodruff, Samson Zhou

Federated Linear Contextual Bandits with User-Level Differential Privacy
Ruiquan Huang, Huanyu Zhang, Luca Melis, Milan Shen, Meisam Hejazinia, Jing Yang

Investigating the Role of Model-Based Learning in Exploration and Transfer
Jacob C Walker, Eszter Vértes, Yazhe Li, Gabriel Dulac-Arnold, Ankesh Anand, Theophane Weber, Jessica B Hamrick

Label Differential Privacy and Private Training Data Release
Robert Busa-Fekete, Andres Munoz, Umar Syed, Sergei Vassilvitskii

Lifelong Language Pretraining with Distribution-Specialized Experts
Wuyang Chen*, Yanqi Zhou, Nan Du, Yanping Huang, James Laudon, Zhifeng Chen, Claire Cui

Multi-User Reinforcement Learning with Low Rank Rewards
Dheeraj Mysore Nagaraj, Suhas S Kowshik, Naman Agarwal, Praneeth Netrapalli, Prateek Jain

Multi-View Masked World Models for Visual Robotic Manipulation
Younggyo Seo, Junsu Kim, Stephen James, Kimin Lee, Jinwoo Shin, Pieter Abbeel

PaLM-E: An Embodied Multimodal Language Model (see blog post)
Danny Driess, Fei Xia, Mehdi S. M. Sajjadi, Corey Lynch, Aakanksha Chowdhery, Brian Ichter,Ayzaan Wahid, Jonathan Tompson, Quan Vuong, Tianhe Yu, Wenlong Huang, Yevgen Chebotar, Pierre Sermanet, Daniel Duckworth, Sergey Levine, Vincent Vanhoucke, Karol Hausman, Marc Toussaint, Klaus Greff, Andy Zeng, Igor Mordatch, Pete Florence

Private Federated Learning with Autotuned Compression
Enayat Ullah*, Christopher A. Choquette-Choo, Peter Kairouz, Sewoong Oh

Refined Regret for Adversarial MDPs with Linear Function Approximation
Yan Dai, Haipeng Luo, Chen-Yu Wei, Julian Zimmert

Scaling Up Dataset Distillation to ImageNet-1K with Constant Memory
Justin Cui, Ruoche Wan, Si Si, Cho-Jui Hsieh

SGD with AdaGrad Stepsizes: Full Adaptivity with High Probability to Unknown Parameters, Unbounded Gradients and Affine Variance
Amit Attia, Tomer Koren

The Statistical Benefits of Quantile Temporal-Difference Learning for Value Estimation
Mark Rowland, Yunhao Tang, Clare Lyle, Rémi Munos, Marc G. Bellemare, Will Dabney

Unveiling The Mask of Position-Information Pattern Through the Mist of Image Features
Chieh Hubert Lin, Hung-Yu Tseng, Hsin-Ying Lee, Maneesh Kumar Singh, Ming-Hsuan Yang

User-Level Private Stochastic Convex Optimization with Optimal Rates
Raef Bassily, Ziteng Sun

A Simple Zero-Shot Prompt Weighting Technique to Improve Prompt Ensembling in Text-Image Models
James Urquhart Allingham*, Jie Ren, Michael W Dusenberry, Xiuye Gu, Yin Cui, Dustin Tran, Jeremiah Zhe Liu, Balaji Lakshminarayanan

Can Large Language Models Reason About Program Invariants?
Kexin Pei, David Bieber, Kensen Shi, Charles Sutton, Pengcheng Yin

Concurrent Shuffle Differential Privacy Under Continual Observation
Jay Tenenbaum, Haim Kaplan, Yishay Mansour, Uri Stemmer

Constant Matters: Fine-Grained Error Bound on Differentially Private Continual Observation
Hendrik Fichtenberger, Monika Henzinger, Jalaj Upadhyay

Cross-Entropy Loss Functions: Theoretical Analysis and Applications
Anqi Mao, Mehryar Mohri, Yutao Zhong

Efficient Rate Optimal Regret for Adversarial Contextual MDPs Using Online Function Approximation
Orin Levy, Alon Cohen, Asaf Cassel, Yishay Mansour

Fairness in Streaming Submodular Maximization Over a Matroid Constraint
Marwa El Halabi, Federico Fusco, Ashkan Norouzi-Fard, Jakab Tardos, Jakub Tarnawski

The Flan Collection: Designing Data and Methods for Effective Instruction Tuning (see blog post)
Shayne Longpre, Le Hou, Tu Vu, Albert Webson, Hyung Won Chung, Yi Tay, Denny Zhou, Quoc V Le, Barret Zoph, Jason Wei, Adam Roberts

Graph Reinforcement Learning for Network Control via Bi-level Optimization
Daniele Gammelli, James Harrison, Kaidi Yang, Marco Pavone, Filipe Rodrigues, Francisco C. Pereira

Learning-Augmented Private Algorithms for Multiple Quantile Release
Mikhail Khodak*, Kareem Amin, Travis Dick, Sergei Vassilvitskii

LegendreTron: Uprising Proper Multiclass Loss Learning
Kevin H Lam, Christian Walder, Spiridon Penev, Richard Nock

Measuring the Impact of Programming Language Distribution
Gabriel Orlanski*, Kefan Xiao, Xavier Garcia, Jeffrey Hui, Joshua Howland, Jonathan Malmaud, Jacob Austin, Rishabh Singh, Michele Catasta*

Multi-task Differential Privacy Under Distribution Skew
Walid Krichene, Prateek Jain, Shuang Song, Mukund Sundararajan, Abhradeep Thakurta, Li Zhang

Muse: Text-to-Image Generation via Masked Generative Transformers
Huiwen Chang, Han Zhang, Jarred Barber, AJ Maschinot, José Lezama, Lu Jiang, Ming-Hsuan Yang, Kevin Murphy, William T. Freeman, Michael Rubinstein, Yuanzhen Li, Dilip Krishnan

On the Convergence of Federated Averaging with Cyclic Client Participation
Yae Jee Cho, Pranay Sharma, Gauri Joshi, Zheng Xu, Satyen Kale, Tong Zhang

Optimal Stochastic Non-smooth Non-convex Optimization Through Online-to-Non-convex Conversion
Ashok Cutkosky, Harsh Mehta, Francesco Orabona

Out-of-Domain Robustness via Targeted Augmentations
Irena Gao, Shiori Sagawa, Pang Wei Koh, Tatsunori Hashimoto, Percy Liang

Polynomial Time and Private Learning of Unbounded Gaussian Mixture Models
Jamil Arbas, Hassan Ashtiani, Christopher Liaw

Pre-computed Memory or On-the-Fly Encoding? A Hybrid Approach to Retrieval Augmentation Makes the Most of Your Compute
Michiel de Jong, Yury Zemlyanskiy, Nicholas FitzGerald, Joshua Ainslie, Sumit Sanghai, Fei Sha, William W. Cohen

Scalable Adaptive Computation for Iterative Generation
Allan Jabri*, David J. Fleet, Ting Chen

Scaling Spherical CNNs
Carlos Esteves, Jean-Jacques Slotine, Ameesh Makadia

STEP: Learning N:M Structured Sparsity Masks from Scratch with Precondition
Yucheng Lu, Shivani Agrawal, Suvinay Subramanian, Oleg Rybakov, Christopher De Sa, Amir Yazdanbakhsh

Stratified Adversarial Robustness with Rejection
Jiefeng Chen, Jayaram Raghuram, Jihye Choi, Xi Wu, Yingyu Liang, Somesh Jha

When Does Privileged information Explain Away Label Noise?
Guillermo Ortiz-Jimenez*, Mark Collier, Anant Nawalgaria, Alexander D'Amour, Jesse Berent, Rodolphe Jenatton, Effrosyni Kokiopoulou

Adaptive Computation with Elastic Input Sequence
Fuzhao Xue*, Valerii Likhosherstov, Anurag Arnab, Neil Houlsby, Mostafa Dehghani, Yang You

Can Neural Network Memorization Be Localized?
Pratyush Maini, Michael C. Mozer, Hanie Sedghi, Zachary C. Lipton, J. Zico Kolter, Chiyuan Zhang

Controllability-Aware Unsupervised Skill Discovery
Seohong Park, Kimin Lee, Youngwoon Lee, Pieter Abbeel

Efficient Learning of Mesh-Based Physical Simulation with Bi-Stride Multi-Scale Graph Neural Network
Yadi Cao, Menglei Chai, Minchen Li, Chenfanfu Jiang

Federated Heavy Hitter Recovery Under Linear Sketching
Adria Gascon, Peter Kairouz, Ziteng Sun, Ananda Theertha Suresh

Graph Generative Model for Benchmarking Graph Neural Networks
Minji Yoon, Yue Wu, John Palowitch, Bryan Perozzi, Russ Salakhutdinov

H-Consistency Bounds for Pairwise Misranking Loss Surrogates
Anqi Mao, Mehryar Mohri, Yutao Zhong

Improved Regret for Efficient Online Reinforcement Learning with Linear Function Approximation
Uri Sherman, Tomer Koren, Yishay Mansour

Invariant Slot Attention: Object Discovery with Slot-Centric Reference Frames
Ondrej Biza*, Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Gamaleldin Fathy Elsayed, Aravindh Mahendran, Thomas Kipf

Multi-task Off-Policy Learning from Bandit Feedback
Joey Hong, Branislav Kveton, Manzil Zaheer, Sumeet Katariya, Mohammad Ghavamzadeh

Optimal No-Regret Learning for One-Sided Lipschitz Functions
Paul Duetting, Guru Guruganesh, Jon Schneider, Joshua Ruizhi Wang

Policy Mirror Ascent for Efficient and Independent Learning in Mean Field Games
Batuhan Yardim, Semih Cayci, Matthieu Geist, Niao He

Regret Minimization and Convergence to Equilibria in General-Sum Markov Games
Liad Erez, Tal Lancewicki, Uri Sherman, Tomer Koren, Yishay Mansour

Reinforcement Learning Can Be More Efficient with Multiple Rewards
Christoph Dann, Yishay Mansour, Mehryar Mohri

Reinforcement Learning with History-Dependent Dynamic Contexts
Guy Tennenholtz, Nadav Merlis, Lior Shani, Martin Mladenov, Craig Boutlier

User-Defined Event Sampling and Uncertainty Quantification in Diffusion Models for Physical Dynamical Systems
Marc Anton Finzi*, Anudhyan Boral, Andrew Gordon Wilson, Fei Sha, Leonardo Zepeda-Nunez

Discrete Key-Value Bottleneck
Frederik Träuble, Anirudh Goyal, Nasim Rahaman, Michael Curtis Mozer, Kenji Kawaguchi, Yoshua Bengio, Bernhard Schölkopf

DSGD-CECA: Decentralized SGD with Communication-Optimal Exact Consensus Algorithm
Lisang Ding, Kexin Jin, Bicheng Ying, Kun Yuan, Wotao Yin

Exphormer: Sparse Transformers for Graphs
Hamed Shirzad, Ameya Velingker, Balaji Venkatachalam, Danica J. Sutherland, Ali Kemal Sinop

Fast, Differentiable and Sparse Top-k: A Convex Analysis Perspective
Michael Eli Sander*, Joan Puigcerver, Josip Djolonga, Gabriel Peyré, Mathieu Blondel

Improved Policy Evaluation for Randomized Trials of Algorithmic Resource Allocation
Aditya Mate, Bryan Wilder, Aparna Taneja, Milind Tambe

In Search for a Generalizable Method for Source Free Domain Adaptation
Malik Boudiaf*, Tom Denton, Bart van Merrienboer, Vincent Dumoulin, Eleni Triantafillou

Learning Rate Schedules in the Presence of Distribution Shift
Matthew Fahrbach, Adel Javanmard, Vahab Mirrokni, Pratik Worah

Not All Semantics Are Created Equal: Contrastive Self-Supervised Learning with Automatic Temperature Individualization
Zi-Hao Qiu, Quanqi Hu, Zhuoning Yuan, Denny Zhou, Lijun Zhang, Tianbao Yang

On the Relationship Between Explanation and Prediction: A Causal View
Amir-Hossein Karimi*, Krikamol Muandet, Simon Kornblith, Bernhard Schölkopf, Been Kim

On the Role of Attention in Prompt-Tuning
Samet Oymak, Ankit Singh Rawat, Mahdi Soltanolkotabi, Christos Thrampoulidis

PLay: Parametrically Conditioned Layout Generation Using Latent Diffusion
Chin-Yi Cheng, Forrest Huang, Gang Li, Yang Li

The Power of Learned Locally Linear Models for Nonlinear Policy Optimization
Daniel Pfrommer, Max Simchowitz, Tyler Westenbroek, Nikolai Matni, Stephen Tu

Relevant Walk Search for Explaining Graph Neural Networks
Ping Xiong, Thomas Schnake, Michael Gastegger, Grégoire Montavon, Klaus Robert Muller,Shinichi Nakajima

Repository-Level Prompt Generation for Large Language Models of Code
Disha Shrivastava, Hugo Larochelle, Daniel Tarlow

Robust and Private Stochastic Linear Bandits
Vasileios Charisopoulos*, Hossein Esfandiari, Vahab Mirrokni

Simple Diffusion: End-to-End Diffusion for High Resolution Images
Emiel Hoogeboom, Jonathan Heek, Tim Salimans

Tied-Augment: Controlling Representation Similarity Improves Data Augmentation
Emirhan Kurtulus, Zichao Li, Yann Dauphin, Ekin D. Cubuk

Why Is Public Pre-Training Necessary for Private Model Training?
Arun Ganesh, Mahdi Haghifam*, Milad Nasr, Sewoong Oh, Thomas Steinke, Om Thakkar, Abhradeep Guha Thakurta, Lun Wang

A Connection Between One-Step RL and Critic Regularization in Reinforcement Learning
Benjamin Eysenbach, Matthieu Geist, Sergey Levine, Ruslan Salakhutdinov

Beyond Uniform Lipschitz Condition in Differentially Private Optimization
Rudrajit Das*, Satyen Kale, Zheng Xu, Tong Zhang, Sujay Sanghavi

Efficient Graph Field Integrators Meet Point Clouds
Krzysztof Choromanski, Arijit Sehanobish, Han Lin, Yunfan Zhao, Eli Berger, Tetiana Parshakova, Alvin Pan, David Watkins, Tianyi Zhang, Valerii Likhosherstov, Somnath Basu Roy Chowdhury, Avinava Dubey, Deepali Jain, Tamas Sarlos, Snigdha Chaturvedi, Adrian Weller

Fast as CHITA: Neural Network Pruning with Combinatorial Optimization
Riade Benbaki, Wenyu Chen, Xiang Meng, Hussein Hazimeh, Natalia Ponomareva, Zhe Zhao, Rahul Mazumder

Jump-Start Reinforcement Learning (see blog post)
Ikechukwu Uchendu*, Ted Xiao, Yao Lu, Banghua Zhu, Mengyuan Yan, Joséphine Simon, Matthew Bennice, Chuyuan Fu, Cong Ma, Jiantao Jiao, Sergey Levine, Karol Hausman

Learning in POMDPs is Sample-Efficient with Hindsight Observability
Jonathan Lee, Alekh Agarwal, Christoph Dann, Tong Zhang

Low-Variance Gradient Estimation in Unrolled Computation Graphs with ES-Single
Paul Vicol

Masked Trajectory Models for Prediction, Representation, and Control
Philipp Wu, Arjun Majumdar, Kevin Stone, Yixin Lin, Igor Mordatch, Pieter Abbeel, Aravind Rajeswaran

Overcoming Simplicity Bias in Deep Networks Using a Feature Sieve
Rishabh Tiwari, Pradeep Shenoy

Pairwise Ranking Losses of Click-Through Rates Prediction for Welfare Maximization in Ad Auctions
Boxiang Lyu, Zhe Feng, Zachary Robertson, Sanmi Koyejo

Predictive Flows for Faster Ford-Fulkerson
Sami Davies, Benjamin Moseley, Sergei Vassilvitskii, Yuyan Wang

Scaling Laws for Multilingual Neural Machine Translation
Patrick Fernandes, Behrooz Ghorbani, Xavier Garcia, Markus Freitag, Orhan Firat

Sequential Monte Carlo Learning for Time Series Structure Discovery
Feras Saad, Brian Patton, Matthew Douglas Hoffman, Rif A. Saurous, Vikash Mansinghka

Stochastic Gradient Succeeds for Bandits
Jincheng Mei, Zixin Zhong, Bo Dai, Alekh Agarwal, Csaba Szepesvari, Dale Schuurmans

Subset-Based Instance Optimality in Private Estimation
Travis Dick, Alex Kulesza, Ziteng Sun, Ananda Theertha Suresh

The Unreasonable Effectiveness of Few-Shot Learning for Machine Translation
Xavier Garcia, Yamini Bansal, Colin Cherry, George Foster, Maxim Krikun, Melvin Johnson, Orhan Firat



Tutorials

Self-Supervised Learning in Vision: from Research Advances to Best Practices
Xinlei Chen, Ishan Misra, Randall Balestriero, Mathilde Caron, Christoph Feichtenhofer, Mark Ibrahim

How to DP-fy ML: A Practical Tutorial to Machine Learning with Differential Privacy (see blog post)
Sergei Vassilvitskii, Natalia Ponomareva, Zheng Xu

Recent Advances in the Generalization Theory of Neural Networks
Tengyu Ma, Alex Damian



EXPO Day workshops

Graph Neural Networks in Tensorflow: A Practical Guide
Workshop Organizers include: Bryan Perozzi, Anton Tsitsulin, Brandon Mayer, Jonathan Halcrow



Google sponsored affinity workshops

LatinX in AI (LAXAI)
Platinum Sponsor
Keynote Speaker: Monica Ribero
Panelist: Yao Qin

Women in Machine Learning (WiML)
Platinum Sponsor
Panelists: Yao Qin



Workshops

Federated Learning and Analytics in Practice: Algorithms, Systems, Applications, and Opportunities
Organizer: Peter Kairouz, Zheng Xu
Speaker: Brendan McMahan

Interpretable Machine Learning in Healthcare (IMLH)
Organizer: Ramin Zabih

Knowledge and Logical Reasoning in the Era of Data-Driven Learning
Organizer: Beliz Günel

The Many Facets of Preference-Based Learning (MFPL)
Organizer: Robert Busa-Fekete, Mohammad Ghavamzadeh

The Synergy of Scientific and Machine Learning Modelling (SynS & ML)
Speaker: Sercan Arik

Theory of Mind in Communicating Agents
Organizer: Pei Zhou

Artificial Intelligence & Human Computer Interaction
Organizer: Yang Li, Forrest Huang

Data-Centric Machine Learning Research (DMLR)
Organizer: Alicia Parrish, Najoung Kim
Speaker: Peter Mattson

Neural Compression: from Information Theory to Applications
Speaker: Johannes Ballé
Panelist: George Toderici

Neural Conversational AI Workshop - What’s Left to TEACH (Trustworthy, Enhanced, Adaptable, Capable and Human-centric) Chatbots?
Organizer: Ahmad Beirami

Spurious Correlations, Invariance and Stability (SCIS)
Organizer: Amir Feder

* Work done while at Google

Source: Google AI Blog


Using societal context knowledge to foster the responsible application of AI

Posted by Donald Martin, Jr., Technical Program Manager, Head of Societal Context Understanding Tools and Solutions (SCOUTS), Google Research

AI-related products and technologies are constructed and deployed in a societal context: that is, a dynamic and complex collection of social, cultural, historical, political and economic circumstances. Because societal contexts by nature are dynamic, complex, non-linear, contested, subjective, and highly qualitative, they are challenging to translate into the quantitative representations, methods, and practices that dominate standard machine learning (ML) approaches and responsible AI product development practices.

The first phase of AI product development is problem understanding, and this phase has tremendous influence over how problems (e.g., increasing cancer screening availability and accuracy) are formulated for ML systems to solve as well many other downstream decisions, such as dataset and ML architecture choice. When the societal context in which a product will operate is not articulated well enough to result in robust problem understanding, the resulting ML solutions can be fragile and even propagate unfair biases.

When AI product developers lack access to the knowledge and tools necessary to effectively understand and consider societal context during development, they tend to abstract it away. This abstraction leaves them with a shallow, quantitative understanding of the problems they seek to solve, while product users and society stakeholders — who are proximate to these problems and embedded in related societal contexts — tend to have a deep qualitative understanding of those same problems. This qualitative–quantitative divergence in ways of understanding complex problems that separates product users and society from developers is what we call the problem understanding chasm.

This chasm has repercussions in the real world: for example, it was the root cause of racial bias discovered by a widely used healthcare algorithm intended to solve the problem of choosing patients with the most complex healthcare needs for special programs. Incomplete understanding of the societal context in which the algorithm would operate led system designers to form incorrect and oversimplified causal theories about what the key problem factors were. Critical socio-structural factors, including lack of access to healthcare, lack of trust in the health care system, and underdiagnosis due to human bias, were left out while spending on healthcare was highlighted as a predictor of complex health need.

To bridge the problem understanding chasm responsibly, AI product developers need tools that put community-validated and structured knowledge of societal context about complex societal problems at their fingertips — starting with problem understanding, but also throughout the product development lifecycle. To that end, Societal Context Understanding Tools and Solutions (SCOUTS) — part of the Responsible AI and Human-Centered Technology (RAI-HCT) team within Google Research — is a dedicated research team focused on the mission to “empower people with the scalable, trustworthy societal context knowledge required to realize responsible, robust AI and solve the world's most complex societal problems.” SCOUTS is motivated by the significant challenge of articulating societal context, and it conducts innovative foundational and applied research to produce structured societal context knowledge and to integrate it into all phases of the AI-related product development lifecycle. Last year we announced that Jigsaw, Google’s incubator for building technology that explores solutions to threats to open societies, leveraged our structured societal context knowledge approach during the data preparation and evaluation phases of model development to scale bias mitigation for their widely used Perspective API toxicity classifier. Going forward SCOUTS’ research agenda focuses on the problem understanding phase of AI-related product development with the goal of bridging the problem understanding chasm.


Bridging the AI problem understanding chasm

Bridging the AI problem understanding chasm requires two key ingredients: 1) a reference frame for organizing structured societal context knowledge and 2) participatory, non-extractive methods to elicit community expertise about complex problems and represent it as structured knowledge. SCOUTS has published innovative research in both areas.




An illustration of the problem understanding chasm.

A societal context reference frame

An essential ingredient for producing structured knowledge is a taxonomy for creating the structure to organize it. SCOUTS collaborated with other RAI-HCT teams (TasC, Impact Lab), Google DeepMind, and external system dynamics experts to develop a taxonomic reference frame for societal context. To contend with the complex, dynamic, and adaptive nature of societal context, we leverage complex adaptive systems (CAS) theory to propose a high-level taxonomic model for organizing societal context knowledge. The model pinpoints three key elements of societal context and the dynamic feedback loops that bind them together: agents, precepts, and artifacts.

  • Agents: These can be individuals or institutions.
  • Precepts: The preconceptions — including beliefs, values, stereotypes and biases — that constrain and drive the behavior of agents. An example of a basic precept is that “all basketball players are over 6 feet tall.” That limiting assumption can lead to failures in identifying basketball players of smaller stature.
  • Artifacts: Agent behaviors produce many kinds of artifacts, including language, data, technologies, societal problems and products.

The relationships between these entities are dynamic and complex. Our work hypothesizes that precepts are the most critical element of societal context and we highlight the problems people perceive and the causal theories they hold about why those problems exist as particularly influential precepts that are core to understanding societal context. For example, in the case of racial bias in a medical algorithm described earlier, the causal theory precept held by designers was that complex health problems would cause healthcare expenditures to go up for all populations. That incorrect precept directly led to the choice of healthcare spending as the proxy variable for the model to predict complex healthcare need, which in turn led to the model being biased against Black patients who, due to societal factors such as lack of access to healthcare and underdiagnosis due to bias on average, do not always spend more on healthcare when they have complex healthcare needs. A key open question is how can we ethically and equitably elicit causal theories from the people and communities who are most proximate to problems of inequity and transform them into useful structured knowledge?

Illustrative version of societal context reference frame.
Taxonomic version of societal context reference frame.

Working with communities to foster the responsible application of AI to healthcare

Since its inception, SCOUTS has worked to build capacity in historically marginalized communities to articulate the broader societal context of the complex problems that matter to them using a practice called community based system dynamics (CBSD). System dynamics (SD) is a methodology for articulating causal theories about complex problems, both qualitatively as causal loop and stock and flow diagrams (CLDs and SFDs, respectively) and quantitatively as simulation models. The inherent support of visual qualitative tools, quantitative methods, and collaborative model building makes it an ideal ingredient for bridging the problem understanding chasm. CBSD is a community-based, participatory variant of SD specifically focused on building capacity within communities to collaboratively describe and model the problems they face as causal theories, directly without intermediaries. With CBSD we’ve witnessed community groups learn the basics and begin drawing CLDs within 2 hours.

Data 4 Black Lives community members learning system dynamics.

There is a huge potential for AI to improve medical diagnosis. But the safety, equity, and reliability of AI-related health diagnostic algorithms depends on diverse and balanced training datasets. An open challenge in the health diagnostic space is the dearth of training sample data from historically marginalized groups. SCOUTS collaborated with the Data 4 Black Lives community and CBSD experts to produce qualitative and quantitative causal theories for the data gap problem. The theories include critical factors that make up the broader societal context surrounding health diagnostics, including cultural memory of death and trust in medical care.

The figure below depicts the causal theory generated during the collaboration described above as a CLD. It hypothesizes that trust in medical care influences all parts of this complex system and is the key lever for increasing screening, which in turn generates data to overcome the data diversity gap.

Causal loop diagram of the health diagnostics data gap

These community-sourced causal theories are a first step to bridge the problem understanding chasm with trustworthy societal context knowledge.


Conclusion

As discussed in this blog, the problem understanding chasm is a critical open challenge in responsible AI. SCOUTS conducts exploratory and applied research in collaboration with other teams within Google Research, external community, and academic partners across multiple disciplines to make meaningful progress solving it. Going forward our work will focus on three key elements, guided by our AI Principles:

  1. Increase awareness and understanding of the problem understanding chasm and its implications through talks, publications, and training.
  2. Conduct foundational and applied research for representing and integrating societal context knowledge into AI product development tools and workflows, from conception to monitoring, evaluation and adaptation.
  3. Apply community-based causal modeling methods to the AI health equity domain to realize impact and build society’s and Google’s capability to produce and leverage global-scale societal context knowledge to realize responsible AI.
SCOUTS flywheel for bridging the problem understanding chasm.

Acknowledgments

Thank you to John Guilyard for graphics development, everyone in SCOUTS, and all of our collaborators and sponsors.

Source: Google AI Blog


SimPer: Simple self-supervised learning of periodic targets

Posted by Daniel McDuff, Staff Research Scientist, and Yuzhe Yang, Student Researcher, Google

Learning from periodic data (signals that repeat, such as a heart beat or the daily temperature changes on Earth’s surface) is crucial for many real-world applications, from monitoring weather systems to detecting vital signs. For example, in the environmental remote sensing domain, periodic learning is often needed to enable nowcasting of environmental changes, such as precipitation patterns or land surface temperature. In the health domain, learning from video measurement has shown to extract (quasi-)periodic vital signs such as atrial fibrillation and sleep apnea episodes.

Approaches like RepNet highlight the importance of these types of tasks, and present a solution that recognizes repetitive activities within a single video. However, these are supervised approaches that require a significant amount of data to capture repetitive activities, all labeled to indicate the number of times an action was repeated. Labeling such data is often challenging and resource-intensive, requiring researchers to manually capture gold-standard temporal measurements that are synchronized with the modality of interest (e.g., video or satellite imagery).

Alternatively, self-supervised learning (SSL) methods (e.g., SimCLR and MoCo v2), which leverage a large amount of unlabeled data to learn representations that capture periodic or quasi-periodic temporal dynamics, have demonstrated success in solving classification tasks. However, they overlook the intrinsic periodicity (i.e., the ability to identify if a frame is part of a periodic process) in data and fail to learn robust representations that capture periodic or frequency attributes. This is because periodic learning exhibits characteristics that are distinct from prevailing learning tasks.

Feature similarity is different in the context of periodic representations as compared to static features (e.g., images). For example, videos that are offset by short time delays or are reversed should be similar to the original sample, whereas videos that have been upsampled or downsampled by a factor x should be different from the original sample by a factor of x.

To address these challenges, in “SimPer: Simple Self-Supervised Learning of Periodic Targets”, published at the eleventh International Conference on Learning Representations (ICLR 2023), we introduced a self-supervised contrastive framework for learning periodic information in data. Specifically, SimPer leverages the temporal properties of periodic targets using temporal self-contrastive learning, where positive and negative samples are obtained through periodicity-invariant and periodicity-variant augmentations from the same input instance. We propose periodic feature similarity that explicitly defines how to measure similarity in the context of periodic learning. Moreover, we design a generalized contrastive loss that extends the classic InfoNCE loss to a soft regression variant that enables contrasting over continuous labels (frequency). Next, we demonstrate that SimPer effectively learns period feature representations compared to state-of-the-art SSL methods, highlighting its intriguing properties including better data efficiency, robustness to spurious correlations, and generalization to distribution shifts. Finally, we are excited to release the SimPer code repo with the research community.


The SimPer framework

SimPer introduces a temporal self-contrastive learning framework. Positive and negative samples are obtained through periodicity-invariant and periodicity-variant augmentations from the same input instance. For temporal video examples, periodicity-invariant changes are cropping, rotation or flipping, whereas periodicity-variant changes involve increasing or decreasing the speed of a video.

To explicitly define how to measure similarity in the context of periodic learning, SimPer proposes periodic feature similarity. This construction allows us to formulate training as a contrastive learning task. A model can be trained with data without any labels and then fine-tuned if necessary to map the learned features to specific frequency values.

Given an input sequence x, we know there’s an underlying associated periodic signal. We then transform x to create a series of speed or frequency altered samples, which changes the underlying periodic target, thus creating different negative views. Although the original frequency is unknown, we effectively devise pseudo- speed or frequency labels for the unlabeled input x.

Conventional similarity measures such as cosine similarity emphasize strict proximity between two feature vectors, and are sensitive to index shifted features (which represent different time stamps), reversed features, and features with changed frequencies. In contrast, periodic feature similarity should be high for samples with small temporal shifts and or reversed indexes, while capturing a continuous similarity change when the feature frequency varies. This can be achieved via a similarity metric in the frequency domain, such as the distance between two Fourier transforms.

To harness the intrinsic continuity of augmented samples in the frequency domain, SimPer designs a generalized contrastive loss that extends the classic InfoNCE loss to a soft regression variant that enables contrasting over continuous labels (frequency). This makes it suitable for regression tasks, where the goal is to recover a continuous signal, such as a heart beat.

SimPer constructs negative views of data through transformations in the frequency domain. The input sequence x has an underlying associated periodic signal. SimPer transforms x to create a series of speed or frequency altered samples, which changes the underlying periodic target, thus creating different negative views. Although the original frequency is unknown, we effectively devise pseudo speed or frequency labels for unlabeled input x (periodicity-variant augmentations τ). SimPer takes transformations that do not change the identity of the input and defines these as periodicity-invariant augmentations σ, thus creating different positive views of the sample. Then, it sends these augmented views to the encoder f, which extracts corresponding features.


Results

To evaluate SimPer's performance, we benchmarked it against state-of-the-art SSL schemes (e.g., SimCLR, MoCo v2, BYOL, CVRL) on a set of six diverse periodic learning datasets for common real-world tasks in human behavior analysis, environmental remote sensing, and healthcare. Specifically, below we present results on heart rate measurement and exercise repetition counting from video. The results show that SimPer outperforms the state-of-the-art SSL schemes across all six datasets, highlighting its superior performance in terms of data efficiency, robustness to spurious correlations, and generalization to unseen targets.

Here we show quantitative results on two representative datasets using SimPer pre-trained using various SSL methods and fine-tuned on the labeled data. First, we pre-train SimPer using the Univ. Bourgogne Franche-Comté Remote PhotoPlethysmoGraphy (UBFC) dataset, a human photoplethysmography and heart rate prediction dataset, and compare its performance to state-of-the-art SSL methods. We observe that SimPer outperforms SimCLR, MoCo v2, BYOL, and CVRL methods. The results on the human action counting dataset, Countix, further confirm the benefits of SimPer over others methods as it notably outperforms the supervised baseline. For the feature evaluation results and performance on other datasets, please refer to the paper.

Results of SimCLR, MoCo v2, BYOL, CVRL and SimPer on the Univ. Bourgogne Franche-Comté Remote PhotoPlethysmoGraphy (UBFC) and Countix datasets. Heart rate and repetition count performance is reported as mean absolute error (MAE).


Conclusion and applications

We present SimPer, a self-supervised contrastive framework for learning periodic information in data. We demonstrate that by combining a temporal self-contrastive learning framework, periodicity-invariant and periodicity-variant augmentations, and continuous periodic feature similarity, SimPer provides an intuitive and flexible approach for learning strong feature representations for periodic signals. Moreover, SimPer can be applied to various fields, ranging from environmental remote sensing to healthcare.


Acknowledgements

We would like to thank Yuzhe Yang, Xin Liu, Ming-Zher Poh, Jiang Wu, Silviu Borac, and Dina Katabi for their contributions to this work.

Source: Google AI Blog


Symbol tuning improves in-context learning in language models

Posted by Jerry Wei, Student Researcher, and Denny Zhou, Principal Scientist, Google Research

A key feature of human intelligence is that humans can learn to perform new tasks by reasoning using only a few examples. Scaling up language models has unlocked a range of new applications and paradigms in machine learning, including the ability to perform challenging reasoning tasks via in-context learning. Language models, however, are still sensitive to the way that prompts are given, indicating that they are not reasoning in a robust manner. For instance, language models often require heavy prompt engineering or phrasing tasks as instructions, and they exhibit unexpected behaviors such as performance on tasks being unaffected even when shown incorrect labels.

In “Symbol tuning improves in-context learning in language models”, we propose a simple fine-tuning procedure that we call symbol tuning, which can improve in-context learning by emphasizing input–label mappings. We experiment with symbol tuning across Flan-PaLM models and observe benefits across various settings.

  • Symbol tuning boosts performance on unseen in-context learning tasks and is much more robust to underspecified prompts, such as those without instructions or without natural language labels.
  • Symbol-tuned models are much stronger at algorithmic reasoning tasks.
  • Finally, symbol-tuned models show large improvements in following flipped-labels presented in-context, meaning that they are more capable of using in-context information to override prior knowledge.
An overview of symbol tuning, where models are fine-tuned on tasks where natural language labels are replaced with arbitrary symbols. Symbol tuning relies on the intuition that when instruction and relevant labels are not available, models must use in-context examples to learn the task.

Motivation

Instruction tuning is a common fine-tuning method that has been shown to improve performance and allow models to better follow in-context examples. One shortcoming, however, is that models are not forced to learn to use the examples because the task is redundantly defined in the evaluation example via instructions and natural language labels. For example, on the left in the figure above, although the examples can help the model understand the task (sentiment analysis), they are not strictly necessary since the model could ignore the examples and just read the instruction that indicates what the task is.

In symbol tuning, the model is fine-tuned on examples where the instructions are removed and natural language labels are replaced with semantically-unrelated labels (e.g., “Foo,” “Bar,” etc.). In this setup, the task is unclear without looking at the in-context examples. For example, on the right in the figure above, multiple in-context examples would be needed to figure out the task. Because symbol tuning teaches the model to reason over the in-context examples, symbol-tuned models should have better performance on tasks that require reasoning between in-context examples and their labels.

Datasets and task types used for symbol tuning.

Symbol-tuning procedure

We selected 22 publicly-available natural language processing (NLP) datasets that we use for our symbol-tuning procedure. These tasks have been widely used in the past, and we only chose classification-type tasks since our method requires discrete labels. We then remap labels to a random label from a set of ~30K arbitrary labels selected from one of three categories: integers, character combinations, and words.

For our experiments, we symbol tune Flan-PaLM, the instruction-tuned variants of PaLM. We use three different sizes of Flan-PaLM models: Flan-PaLM-8B, Flan-PaLM-62B, and Flan-PaLM-540B. We also tested Flan-cont-PaLM-62B (Flan-PaLM-62B at 1.3T tokens instead of 780B tokens), which we abbreviate as 62B-c.

We use a set of ∼300K arbitrary symbols from three categories (integers, character combinations, and words). ∼30K symbols are used during tuning and the rest are held out for evaluation.

Experimental setup

We want to evaluate a model’s ability to perform unseen tasks, so we cannot evaluate on tasks used in symbol tuning (22 datasets) or used during instruction tuning (1.8K tasks). Hence, we choose 11 NLP datasets that were not used during fine-tuning.


In-context learning

In the symbol-tuning procedure, models must learn to reason with in-context examples in order to successfully perform tasks because prompts are modified to ensure that tasks cannot simply be learned from relevant labels or instructions. Symbol-tuned models should perform better in settings where tasks are unclear and require reasoning between in-context examples and their labels. To explore these settings, we define four in-context learning settings that vary the amount of reasoning required between inputs and labels in order to learn the task (based on the availability of instructions/relevant labels)

Depending on the availability of instructions and relevant natural language labels, models may need to do varying amounts of reasoning with in-context examples. When these features are not available, models must reason with the given in-context examples to successfully perform the task.

Symbol tuning improves performance across all settings for models 62B and larger, with small improvements in settings with relevant natural language labels (+0.8% to +4.2%) and substantial improvements in settings without relevant natural language labels (+5.5% to +15.5%). Strikingly, when relevant labels are unavailable, symbol-tuned Flan-PaLM-8B outperforms FlanPaLM-62B, and symbol-tuned Flan-PaLM-62B outperforms Flan-PaLM-540B. This performance difference suggests that symbol tuning can allow much smaller models to perform as well as large models on these tasks (effectively saving ∼10X inference compute).

Large-enough symbol-tuned models are better at in-context learning than baselines, especially in settings where relevant labels are not available. Performance is shown as average model accuracy (%) across eleven tasks.

Algorithmic reasoning

We also experiment on algorithmic reasoning tasks from BIG-Bench. There are two main groups of tasks: 1) List functions — identify a transformation function (e.g., remove the last element in a list) between input and output lists containing non-negative integers; and 2) simple turing concepts — reason with binary strings to learn the concept that maps an input to an output (e.g., swapping 0s and 1s in a string).

On the list function and simple turing concept tasks, symbol tuning results in an average performance improvement of 18.2% and 15.3%, respectively. Additionally, Flan-cont-PaLM-62B with symbol tuning outperforms Flan-PaLM-540B on the list function tasks on average, which is equivalent to a ∼10x reduction in inference compute. These improvements suggest that symbol tuning strengthens the model’s ability to learn in-context for unseen task types, as symbol tuning did not include any algorithmic data.

Symbol-tuned models achieve higher performance on list function tasks and simple turing concept tasks. (A–E): categories of list functions tasks. (F): simple turing concepts task.

Flipped labels

In the flipped-label experiment, labels of in-context and evaluation examples are flipped, meaning that prior knowledge and input-label mappings disagree (e.g., sentences containing positive sentiment labeled as “negative sentiment”), thereby allowing us to study whether models can override prior knowledge. Previous work has shown that while pre-trained models (without instruction tuning) can, to some extent, follow flipped labels presented in-context, instruction tuning degraded this ability.

We see that there is a similar trend across all model sizes — symbol-tuned models are much more capable of following flipped labels than instruction-tuned models. We found that after symbol tuning, Flan-PaLM-8B sees an average improvement across all datasets of 26.5%, Flan-PaLM-62B sees an improvement of 33.7%, and Flan-PaLM-540B sees an improvement of 34.0%. Additionally, symbol-tuned models achieve similar or better than average performance as pre-training–only models.

Symbol-tuned models are much better at following flipped labels presented in-context than instruction-tuned models are.

Conclusion

We presented symbol tuning, a new method of tuning models on tasks where natural language labels are remapped to arbitrary symbols. Symbol tuning is based off of the intuition that when models cannot use instructions or relevant labels to determine a presented task, it must do so by instead learning from in-context examples. We tuned four language models using our symbol-tuning procedure, utilizing a tuning mixture of 22 datasets and approximately 30K arbitrary symbols as labels.

We first showed that symbol tuning improves performance on unseen in-context learning tasks, especially when prompts do not contain instructions or relevant labels. We also found that symbol-tuned models were much better at algorithmic reasoning tasks, despite the lack of numerical or algorithmic data in the symbol-tuning procedure. Finally, in an in-context learning setting where inputs have flipped labels, symbol tuning (for some datasets) restores the ability to follow flipped labels that was lost during instruction tuning.


Future work

Through symbol tuning, we aim to increase the degree to which models can examine and learn from input–label mappings during in-context learning. We hope that our results encourage further work towards improving language models’ ability to reason over symbols presented in-context.


Acknowledgements

The authors of this post are now part of Google DeepMind. This work was conducted by Jerry Wei, Le Hou, Andrew Lampinen, Xiangning Chen, Da Huang, Yi Tay, Xinyun Chen, Yifeng Lu, Denny Zhou, Tengyu Ma, and Quoc V. Le. We would like to thank our colleagues at Google Research and Google DeepMind for their advice and helpful discussions.

Source: Google AI Blog


Symbol tuning improves in-context learning in language models

Posted by Jerry Wei, Student Researcher, and Denny Zhou, Principal Scientist, Google Research

A key feature of human intelligence is that humans can learn to perform new tasks by reasoning using only a few examples. Scaling up language models has unlocked a range of new applications and paradigms in machine learning, including the ability to perform challenging reasoning tasks via in-context learning. Language models, however, are still sensitive to the way that prompts are given, indicating that they are not reasoning in a robust manner. For instance, language models often require heavy prompt engineering or phrasing tasks as instructions, and they exhibit unexpected behaviors such as performance on tasks being unaffected even when shown incorrect labels.

In “Symbol tuning improves in-context learning in language models”, we propose a simple fine-tuning procedure that we call symbol tuning, which can improve in-context learning by emphasizing input–label mappings. We experiment with symbol tuning across Flan-PaLM models and observe benefits across various settings.

  • Symbol tuning boosts performance on unseen in-context learning tasks and is much more robust to underspecified prompts, such as those without instructions or without natural language labels.
  • Symbol-tuned models are much stronger at algorithmic reasoning tasks.
  • Finally, symbol-tuned models show large improvements in following flipped-labels presented in-context, meaning that they are more capable of using in-context information to override prior knowledge.
An overview of symbol tuning, where models are fine-tuned on tasks where natural language labels are replaced with arbitrary symbols. Symbol tuning relies on the intuition that when instruction and relevant labels are not available, models must use in-context examples to learn the task.

Motivation

Instruction tuning is a common fine-tuning method that has been shown to improve performance and allow models to better follow in-context examples. One shortcoming, however, is that models are not forced to learn to use the examples because the task is redundantly defined in the evaluation example via instructions and natural language labels. For example, on the left in the figure above, although the examples can help the model understand the task (sentiment analysis), they are not strictly necessary since the model could ignore the examples and just read the instruction that indicates what the task is.

In symbol tuning, the model is fine-tuned on examples where the instructions are removed and natural language labels are replaced with semantically-unrelated labels (e.g., “Foo,” “Bar,” etc.). In this setup, the task is unclear without looking at the in-context examples. For example, on the right in the figure above, multiple in-context examples would be needed to figure out the task. Because symbol tuning teaches the model to reason over the in-context examples, symbol-tuned models should have better performance on tasks that require reasoning between in-context examples and their labels.

Datasets and task types used for symbol tuning.

Symbol-tuning procedure

We selected 22 publicly-available natural language processing (NLP) datasets that we use for our symbol-tuning procedure. These tasks have been widely used in the past, and we only chose classification-type tasks since our method requires discrete labels. We then remap labels to a random label from a set of ~30K arbitrary labels selected from one of three categories: integers, character combinations, and words.

For our experiments, we symbol tune Flan-PaLM, the instruction-tuned variants of PaLM. We use three different sizes of Flan-PaLM models: Flan-PaLM-8B, Flan-PaLM-62B, and Flan-PaLM-540B. We also tested Flan-cont-PaLM-62B (Flan-PaLM-62B at 1.3T tokens instead of 780B tokens), which we abbreviate as 62B-c.

We use a set of ∼300K arbitrary symbols from three categories (integers, character combinations, and words). ∼30K symbols are used during tuning and the rest are held out for evaluation.

Experimental setup

We want to evaluate a model’s ability to perform unseen tasks, so we cannot evaluate on tasks used in symbol tuning (22 datasets) or used during instruction tuning (1.8K tasks). Hence, we choose 11 NLP datasets that were not used during fine-tuning.


In-context learning

In the symbol-tuning procedure, models must learn to reason with in-context examples in order to successfully perform tasks because prompts are modified to ensure that tasks cannot simply be learned from relevant labels or instructions. Symbol-tuned models should perform better in settings where tasks are unclear and require reasoning between in-context examples and their labels. To explore these settings, we define four in-context learning settings that vary the amount of reasoning required between inputs and labels in order to learn the task (based on the availability of instructions/relevant labels)

Depending on the availability of instructions and relevant natural language labels, models may need to do varying amounts of reasoning with in-context examples. When these features are not available, models must reason with the given in-context examples to successfully perform the task.

Symbol tuning improves performance across all settings for models 62B and larger, with small improvements in settings with relevant natural language labels (+0.8% to +4.2%) and substantial improvements in settings without relevant natural language labels (+5.5% to +15.5%). Strikingly, when relevant labels are unavailable, symbol-tuned Flan-PaLM-8B outperforms FlanPaLM-62B, and symbol-tuned Flan-PaLM-62B outperforms Flan-PaLM-540B. This performance difference suggests that symbol tuning can allow much smaller models to perform as well as large models on these tasks (effectively saving ∼10X inference compute).

Large-enough symbol-tuned models are better at in-context learning than baselines, especially in settings where relevant labels are not available. Performance is shown as average model accuracy (%) across eleven tasks.

Algorithmic reasoning

We also experiment on algorithmic reasoning tasks from BIG-Bench. There are two main groups of tasks: 1) List functions — identify a transformation function (e.g., remove the last element in a list) between input and output lists containing non-negative integers; and 2) simple turing concepts — reason with binary strings to learn the concept that maps an input to an output (e.g., swapping 0s and 1s in a string).

On the list function and simple turing concept tasks, symbol tuning results in an average performance improvement of 18.2% and 15.3%, respectively. Additionally, Flan-cont-PaLM-62B with symbol tuning outperforms Flan-PaLM-540B on the list function tasks on average, which is equivalent to a ∼10x reduction in inference compute. These improvements suggest that symbol tuning strengthens the model’s ability to learn in-context for unseen task types, as symbol tuning did not include any algorithmic data.

Symbol-tuned models achieve higher performance on list function tasks and simple turing concept tasks. (A–E): categories of list functions tasks. (F): simple turing concepts task.

Flipped labels

In the flipped-label experiment, labels of in-context and evaluation examples are flipped, meaning that prior knowledge and input-label mappings disagree (e.g., sentences containing positive sentiment labeled as “negative sentiment”), thereby allowing us to study whether models can override prior knowledge. Previous work has shown that while pre-trained models (without instruction tuning) can, to some extent, follow flipped labels presented in-context, instruction tuning degraded this ability.

We see that there is a similar trend across all model sizes — symbol-tuned models are much more capable of following flipped labels than instruction-tuned models. We found that after symbol tuning, Flan-PaLM-8B sees an average improvement across all datasets of 26.5%, Flan-PaLM-62B sees an improvement of 33.7%, and Flan-PaLM-540B sees an improvement of 34.0%. Additionally, symbol-tuned models achieve similar or better than average performance as pre-training–only models.

Symbol-tuned models are much better at following flipped labels presented in-context than instruction-tuned models are.

Conclusion

We presented symbol tuning, a new method of tuning models on tasks where natural language labels are remapped to arbitrary symbols. Symbol tuning is based off of the intuition that when models cannot use instructions or relevant labels to determine a presented task, it must do so by instead learning from in-context examples. We tuned four language models using our symbol-tuning procedure, utilizing a tuning mixture of 22 datasets and approximately 30K arbitrary symbols as labels.

We first showed that symbol tuning improves performance on unseen in-context learning tasks, especially when prompts do not contain instructions or relevant labels. We also found that symbol-tuned models were much better at algorithmic reasoning tasks, despite the lack of numerical or algorithmic data in the symbol-tuning procedure. Finally, in an in-context learning setting where inputs have flipped labels, symbol tuning (for some datasets) restores the ability to follow flipped labels that was lost during instruction tuning.


Future work

Through symbol tuning, we aim to increase the degree to which models can examine and learn from input–label mappings during in-context learning. We hope that our results encourage further work towards improving language models’ ability to reason over symbols presented in-context.


Acknowledgements

The authors of this post are now part of Google DeepMind. This work was conducted by Jerry Wei, Le Hou, Andrew Lampinen, Xiangning Chen, Da Huang, Yi Tay, Xinyun Chen, Yifeng Lu, Denny Zhou, Tengyu Ma, and Quoc V. Le. We would like to thank our colleagues at Google Research and Google DeepMind for their advice and helpful discussions.

Source: Google AI Blog


Symbol tuning improves in-context learning in language models

Posted by Jerry Wei, Student Researcher, and Denny Zhou, Principal Scientist, Google Research

A key feature of human intelligence is that humans can learn to perform new tasks by reasoning using only a few examples. Scaling up language models has unlocked a range of new applications and paradigms in machine learning, including the ability to perform challenging reasoning tasks via in-context learning. Language models, however, are still sensitive to the way that prompts are given, indicating that they are not reasoning in a robust manner. For instance, language models often require heavy prompt engineering or phrasing tasks as instructions, and they exhibit unexpected behaviors such as performance on tasks being unaffected even when shown incorrect labels.

In “Symbol tuning improves in-context learning in language models”, we propose a simple fine-tuning procedure that we call symbol tuning, which can improve in-context learning by emphasizing input–label mappings. We experiment with symbol tuning across Flan-PaLM models and observe benefits across various settings.

  • Symbol tuning boosts performance on unseen in-context learning tasks and is much more robust to underspecified prompts, such as those without instructions or without natural language labels.
  • Symbol-tuned models are much stronger at algorithmic reasoning tasks.
  • Finally, symbol-tuned models show large improvements in following flipped-labels presented in-context, meaning that they are more capable of using in-context information to override prior knowledge.
An overview of symbol tuning, where models are fine-tuned on tasks where natural language labels are replaced with arbitrary symbols. Symbol tuning relies on the intuition that when instruction and relevant labels are not available, models must use in-context examples to learn the task.

Motivation

Instruction tuning is a common fine-tuning method that has been shown to improve performance and allow models to better follow in-context examples. One shortcoming, however, is that models are not forced to learn to use the examples because the task is redundantly defined in the evaluation example via instructions and natural language labels. For example, on the left in the figure above, although the examples can help the model understand the task (sentiment analysis), they are not strictly necessary since the model could ignore the examples and just read the instruction that indicates what the task is.

In symbol tuning, the model is fine-tuned on examples where the instructions are removed and natural language labels are replaced with semantically-unrelated labels (e.g., “Foo,” “Bar,” etc.). In this setup, the task is unclear without looking at the in-context examples. For example, on the right in the figure above, multiple in-context examples would be needed to figure out the task. Because symbol tuning teaches the model to reason over the in-context examples, symbol-tuned models should have better performance on tasks that require reasoning between in-context examples and their labels.

Datasets and task types used for symbol tuning.

Symbol-tuning procedure

We selected 22 publicly-available natural language processing (NLP) datasets that we use for our symbol-tuning procedure. These tasks have been widely used in the past, and we only chose classification-type tasks since our method requires discrete labels. We then remap labels to a random label from a set of ~30K arbitrary labels selected from one of three categories: integers, character combinations, and words.

For our experiments, we symbol tune Flan-PaLM, the instruction-tuned variants of PaLM. We use three different sizes of Flan-PaLM models: Flan-PaLM-8B, Flan-PaLM-62B, and Flan-PaLM-540B. We also tested Flan-cont-PaLM-62B (Flan-PaLM-62B at 1.3T tokens instead of 780B tokens), which we abbreviate as 62B-c.

We use a set of ∼300K arbitrary symbols from three categories (integers, character combinations, and words). ∼30K symbols are used during tuning and the rest are held out for evaluation.

Experimental setup

We want to evaluate a model’s ability to perform unseen tasks, so we cannot evaluate on tasks used in symbol tuning (22 datasets) or used during instruction tuning (1.8K tasks). Hence, we choose 11 NLP datasets that were not used during fine-tuning.


In-context learning

In the symbol-tuning procedure, models must learn to reason with in-context examples in order to successfully perform tasks because prompts are modified to ensure that tasks cannot simply be learned from relevant labels or instructions. Symbol-tuned models should perform better in settings where tasks are unclear and require reasoning between in-context examples and their labels. To explore these settings, we define four in-context learning settings that vary the amount of reasoning required between inputs and labels in order to learn the task (based on the availability of instructions/relevant labels)

Depending on the availability of instructions and relevant natural language labels, models may need to do varying amounts of reasoning with in-context examples. When these features are not available, models must reason with the given in-context examples to successfully perform the task.

Symbol tuning improves performance across all settings for models 62B and larger, with small improvements in settings with relevant natural language labels (+0.8% to +4.2%) and substantial improvements in settings without relevant natural language labels (+5.5% to +15.5%). Strikingly, when relevant labels are unavailable, symbol-tuned Flan-PaLM-8B outperforms FlanPaLM-62B, and symbol-tuned Flan-PaLM-62B outperforms Flan-PaLM-540B. This performance difference suggests that symbol tuning can allow much smaller models to perform as well as large models on these tasks (effectively saving ∼10X inference compute).

Large-enough symbol-tuned models are better at in-context learning than baselines, especially in settings where relevant labels are not available. Performance is shown as average model accuracy (%) across eleven tasks.

Algorithmic reasoning

We also experiment on algorithmic reasoning tasks from BIG-Bench. There are two main groups of tasks: 1) List functions — identify a transformation function (e.g., remove the last element in a list) between input and output lists containing non-negative integers; and 2) simple turing concepts — reason with binary strings to learn the concept that maps an input to an output (e.g., swapping 0s and 1s in a string).

On the list function and simple turing concept tasks, symbol tuning results in an average performance improvement of 18.2% and 15.3%, respectively. Additionally, Flan-cont-PaLM-62B with symbol tuning outperforms Flan-PaLM-540B on the list function tasks on average, which is equivalent to a ∼10x reduction in inference compute. These improvements suggest that symbol tuning strengthens the model’s ability to learn in-context for unseen task types, as symbol tuning did not include any algorithmic data.

Symbol-tuned models achieve higher performance on list function tasks and simple turing concept tasks. (A–E): categories of list functions tasks. (F): simple turing concepts task.

Flipped labels

In the flipped-label experiment, labels of in-context and evaluation examples are flipped, meaning that prior knowledge and input-label mappings disagree (e.g., sentences containing positive sentiment labeled as “negative sentiment”), thereby allowing us to study whether models can override prior knowledge. Previous work has shown that while pre-trained models (without instruction tuning) can, to some extent, follow flipped labels presented in-context, instruction tuning degraded this ability.

We see that there is a similar trend across all model sizes — symbol-tuned models are much more capable of following flipped labels than instruction-tuned models. We found that after symbol tuning, Flan-PaLM-8B sees an average improvement across all datasets of 26.5%, Flan-PaLM-62B sees an improvement of 33.7%, and Flan-PaLM-540B sees an improvement of 34.0%. Additionally, symbol-tuned models achieve similar or better than average performance as pre-training–only models.

Symbol-tuned models are much better at following flipped labels presented in-context than instruction-tuned models are.

Conclusion

We presented symbol tuning, a new method of tuning models on tasks where natural language labels are remapped to arbitrary symbols. Symbol tuning is based off of the intuition that when models cannot use instructions or relevant labels to determine a presented task, it must do so by instead learning from in-context examples. We tuned four language models using our symbol-tuning procedure, utilizing a tuning mixture of 22 datasets and approximately 30K arbitrary symbols as labels.

We first showed that symbol tuning improves performance on unseen in-context learning tasks, especially when prompts do not contain instructions or relevant labels. We also found that symbol-tuned models were much better at algorithmic reasoning tasks, despite the lack of numerical or algorithmic data in the symbol-tuning procedure. Finally, in an in-context learning setting where inputs have flipped labels, symbol tuning (for some datasets) restores the ability to follow flipped labels that was lost during instruction tuning.


Future work

Through symbol tuning, we aim to increase the degree to which models can examine and learn from input–label mappings during in-context learning. We hope that our results encourage further work towards improving language models’ ability to reason over symbols presented in-context.


Acknowledgements

The authors of this post are now part of Google DeepMind. This work was conducted by Jerry Wei, Le Hou, Andrew Lampinen, Xiangning Chen, Da Huang, Yi Tay, Xinyun Chen, Yifeng Lu, Denny Zhou, Tengyu Ma, and Quoc V. Le. We would like to thank our colleagues at Google Research and Google DeepMind for their advice and helpful discussions.

Source: Google AI Blog