Many of the challenges facing today's reinforcement learning (RL) algorithms, such as robustness, generalization, transfer, and computational efficiency are closely related to compression. Prior work has convincingly argued why minimizing information is useful in the supervised learning setting, but standard RL algorithms lack an explicit mechanism for compression. The RL setting is unique because (1) its sequential nature allows an agent to use past information to avoid looking at future observations and (2) the agent can optimize its behavior to prefer states where decision making requires few bits. We take advantage of these properties to propose a method (RPC) for learning simple policies. This method brings together ideas from information bottlenecks, model-based RL, and bits-back coding into a simple and theoretically-justified algorithm. Our method jointly optimizes a latent-space model and policy to be self-consistent, such that the policy avoids states where the model is inaccurate. We demonstrate that our method achieves much tighter compression than prior methods, achieving up to 5x higher reward than a standard information bottleneck. We also demonstrate that our method learns policies that are more robust and generalize better to new tasks.
Learning multimodal representations involves integrating information from multiple heterogeneous sources of data. It is a challenging yet crucial area with numerous real-world applications in multimedia, affective computing, robotics, finance, human-computer interaction, and healthcare. Unfortunately, multimodal research has seen limited resources to study (1) generalization across domains and modalities, (2) complexity during training and inference, and (3) robustness to noisy and missing modalities. In order to accelerate progress towards understudied modalities and tasks while ensuring real-world robustness, we release MultiBench, a systematic and unified large-scale benchmark spanning 15 datasets, 10 modalities, 20 prediction tasks, and 6 research areas. MultiBench provides an automated end-to-end machine learning pipeline that simplifies and standardizes data loading, experimental setup, and model evaluation. To enable holistic evaluation, MultiBench offers a comprehensive methodology to assess (1) generalization, (2) time and space complexity, and (3) modality robustness. MultiBench introduces impactful challenges for future research, including scalability to large-scale multimodal datasets and robustness to realistic imperfections. To accompany this benchmark, we also provide a standardized implementation of 20 core approaches in multimodal learning. Simply applying methods proposed in different research areas can improve the state-of-the-art performance on 9/15 datasets. Therefore, MultiBench presents a milestone in unifying disjoint efforts in multimodal research and paves the way towards a better understanding of the capabilities and limitations of multimodal models, all the while ensuring ease of use, accessibility, and reproducibility. MultiBench, our standardized code, and leaderboards are publicly available, will be regularly updated, and welcomes inputs from the community.
As machine learning methods are deployed in real-world settings such as healthcare, legal systems, and social science, it is crucial to recognize how they shape social biases and stereotypes in these sensitive decision-making processes. Among such real-world deployments are large-scale pretrained language models (LMs) that can be potentially dangerous in manifesting undesirable representational biases - harmful biases resulting from stereotyping that propagate negative generalizations involving gender, race, religion, and other social constructs. As a step towards improving the fairness of LMs, we carefully define several sources of representational biases before proposing new benchmarks and metrics to measure them. With these tools, we propose steps towards mitigating social biases during text generation. Our empirical results and human evaluation demonstrate effectiveness in mitigating bias while retaining crucial contextual information for high-fidelity text generation, thereby pushing forward the performance-fairness Pareto frontier.
Mental health conditions remain underdiagnosed even in countries with common access to advanced medical care. The ability to accurately and efficiently predict mood from easily collectible data has several important implications for the early detection, intervention, and treatment of mental health disorders. One promising data source to help monitor human behavior is daily smartphone usage. However, care must be taken to summarize behaviors without identifying the user through personal (e.g., personally identifiable information) or protected (e.g., race, gender) attributes. In this paper, we study behavioral markers of daily mood using a recent dataset of mobile behaviors from adolescent populations at high risk of suicidal behaviors. Using computational models, we find that language and multimodal representations of mobile typed text (spanning typed characters, words, keystroke timings, and app usage) are predictive of daily mood. However, we find that models trained to predict mood often also capture private user identities in their intermediate representations. To tackle this problem, we evaluate approaches that obfuscate user identity while remaining predictive. By combining multimodal representations with privacy-preserving learning, we are able to push forward the performance-privacy frontier.
Self-supervised approaches for speech representation learning are challenged by three unique problems: (1) there are multiple sound units in each input utterance, (2) there is no lexicon of input sound units during the pre-training phase, and (3) sound units have variable lengths with no explicit segmentation. To deal with these three problems, we propose the Hidden-Unit BERT (HuBERT) approach for self-supervised speech representation learning, which utilizes an offline clustering step to provide aligned target labels for a BERT-like prediction loss. A key ingredient of our approach is applying the prediction loss over the masked regions only, which forces the model to learn a combined acoustic and language model over the continuous inputs. HuBERT relies primarily on the consistency of the unsupervised clustering step rather than the intrinsic quality of the assigned cluster labels. Starting with a simple k-means teacher of 100 clusters, and using two iterations of clustering, the HuBERT model either matches or improves upon the state-of-the-art wav2vec 2.0 performance on the Librispeech (960h) and Libri-light (60,000h) benchmarks with 10min, 1h, 10h, 100h, and 960h fine-tuning subsets. Using a 1B parameter model, HuBERT shows up to 19% and 13% relative WER reduction on the more challenging dev-other and test-other evaluation subsets.
Designing provably efficient algorithms with general function approximation is an important open problem in reinforcement learning. Recently, Wang et al.~[2020c] establish a value-based algorithm with general function approximation that enjoys $\widetilde{O}(\mathrm{poly}(dH)\sqrt{K})$\footnote{Throughout the paper, we use $\widetilde{O}(\cdot)$ to suppress logarithm factors. } regret bound, where $d$ depends on the complexity of the function class, $H$ is the planning horizon, and $K$ is the total number of episodes. However, their algorithm requires $\Omega(K)$ computation time per round, rendering the algorithm inefficient for practical use. In this paper, by applying online sub-sampling techniques, we develop an algorithm that takes $\widetilde{O}(\mathrm{poly}(dH))$ computation time per round on average, and enjoys nearly the same regret bound. Furthermore, the algorithm achieves low switching cost, i.e., it changes the policy only $\widetilde{O}(\mathrm{poly}(dH))$ times during its execution, making it appealing to be implemented in real-life scenarios. Moreover, by using an upper-confidence based exploration-driven reward function, the algorithm provably explores the environment in the reward-free setting. In particular, after $\widetilde{O}(\mathrm{poly}(dH))/\epsilon^2$ rounds of exploration, the algorithm outputs an $\epsilon$-optimal policy for any given reward function.
This paper presents to integrate the auxiliary information (e.g., additional attributes for data such as the hashtags for Instagram images) in the self-supervised learning process. We first observe that the auxiliary information may bring us useful information about data structures: for instance, the Instagram images with the same hashtags can be semantically similar. Hence, to leverage the structural information from the auxiliary information, we present to construct data clusters according to the auxiliary information. Then, we introduce the Clustering InfoNCE (Cl-InfoNCE) objective that learns similar representations for augmented variants of data from the same cluster and dissimilar representations for data from different clusters. Our approach contributes as follows: 1) Comparing to conventional self-supervised representations, the auxiliary-information-infused self-supervised representations bring the performance closer to the supervised representations; 2) The presented Cl-InfoNCE can also work with unsupervised constructed clusters (e.g., k-means clusters) and outperform strong clustering-based self-supervised learning approaches, such as the Prototypical Contrastive Learning (PCL) method; 3) We show that Cl-InfoNCE may be a better approach to leverage the data clustering information, by comparing it to the baseline approach - learning to predict the clustering assignments with cross-entropy loss. For analysis, we connect the goodness of the learned representations with the statistical relationships: i) the mutual information between the labels and the clusters and ii) the conditional entropy of the clusters given the labels.
Self-supervised learning is a form of unsupervised learning that leverages rich information in data to learn representations. However, data sometimes contains certain information that may be undesirable for downstream tasks. For instance, gender information may lead to biased decisions on many gender-irrelevant tasks. In this paper, we develop conditional contrastive learning to remove undesirable information in self-supervised representations. To remove the effect of the undesirable variable, our proposed approach conditions on the undesirable variable (i.e., by fixing the variations of it) during the contrastive learning process. In particular, inspired by the contrastive objective InfoNCE, we introduce Conditional InfoNCE (C-InfoNCE), and its computationally efficient variant, Weak-Conditional InfoNCE (WeaC-InfoNCE), for conditional contrastive learning. We demonstrate empirically that our methods can successfully learn self-supervised representations for downstream tasks while removing a great level of information related to the undesirable variables. We study three scenarios, each with a different type of undesirable variables: task-irrelevant meta-information for self-supervised speech representation learning, sensitive attributes for fair representation learning, and domain specification for multi-domain visual representation learning.
Answering complex questions from long documents requires aggregating multiple pieces of evidence and then predicting the answers. In this paper, we propose a multi-hop retrieval method, DocHopper, to answer compositional questions over long documents. At each step, DocHopper retrieves a paragraph or sentence embedding from the document, mixes the retrieved result with the query, and updates the query for the next step. In contrast to many other retrieval-based methods (e.g., RAG or REALM) the query is not augmented with a token sequence: instead, it is augmented by "numerically" combining it with another neural representation. This means that model is end-to-end differentiable. We demonstrate that utilizing document structure in this was can largely improve question-answering and retrieval performance on long documents. We experimented with DocHopper on three different QA tasks that require reading long documents to answer compositional questions: discourse entailment reasoning, factual QA with table and text, and information seeking QA from academic papers. DocHopper outperforms all baseline models and achieves state-of-the-art results on all datasets. Additionally, DocHopper is efficient at inference time, being 3~10 times faster than the baselines.
Offline Reinforcement Learning promises to learn effective policies from previously-collected, static datasets without the need for exploration. However, existing Q-learning and actor-critic based off-policy RL algorithms fail when bootstrapping from out-of-distribution (OOD) actions or states. We hypothesize that a key missing ingredient from the existing methods is a proper treatment of uncertainty in the offline setting. We propose Uncertainty Weighted Actor-Critic (UWAC), an algorithm that detects OOD state-action pairs and down-weights their contribution in the training objectives accordingly. Implementation-wise, we adopt a practical and effective dropout-based uncertainty estimation method that introduces very little overhead over existing RL algorithms. Empirically, we observe that UWAC substantially improves model stability during training. In addition, UWAC out-performs existing offline RL methods on a variety of competitive tasks, and achieves significant performance gains over the state-of-the-art baseline on datasets with sparse demonstrations collected from human experts.