Publications

A Consciousness-Inspired Planning Agent for Model-Based Reinforcement Learning

Mingde Zhao

We present an end-to-end, model-based deep reinforcement learning agent which dynamically attends to relevant parts of its state during plan… (voir plus)ning. The agent uses a bottleneck mechanism over a set-based representation to force the number of entities to which the agent attends at each planning step to be small. In experiments, we investigate the bottleneck mechanism with several sets of customized environments featuring different challenges. We consistently observe that the design allows the planning agents to generalize their learned task-solving abilities in compatible unseen environments by attending to the relevant objects, leading to better out-of-distribution generalization performance.

2021-12-06

Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

Continual Learning via Local Module Composition

Oleksiy Ostapenko

Pau Rodríguez

Massimo Caccia

Laurent Charlin

Modularity is a compelling solution to continual learning (CL), the problem of modeling sequences of related tasks. Learning and then compos… (voir plus)ing modules to solve different tasks provides an abstraction to address the principal challenges of CL including catastrophic forgetting, backward and forward transfer across tasks, and sub-linear model growth. We introduce local module composition (LMC), an approach to modular CL where each module is provided a local structural component that estimates a module's relevance to the input. Dynamic module composition is performed layer-wise based on local relevance scores. We demonstrate that agnosticity to task identities (IDs) arises from (local) structural learning that is module-specific as opposed to the task- and/or model-specific as in previous works, making LMC applicable to more CL settings compared to previous works. In addition, LMC also tracks statistics about the input distribution and adds new modules when outlier samples are detected. In the first set of experiments, LMC performs favorably compared to existing methods on the recent Continual Transfer-learning Benchmark without requiring task identities. In another study, we show that the locality of structural learning allows LMC to interpolate to related but unseen tasks (OOD), as well as to compose modular networks trained independently on different task sequences into a third modular network without any fine-tuning. Finally, in search for limitations of LMC we study it on more challenging sequences of 30 and 100 tasks, demonstrating that local module selection becomes much more challenging in presence of a large number of candidate modules. In this setting best performing LMC spawns much fewer modules compared to an oracle based baseline, however, it reaches a lower overall accuracy. The codebase is available under https://github.com/oleksost/LMC.

2021-12-06

Conference on Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

Curating the Twitter Election Integrity Datasets for Better Online Troll Characterization

Albert M. Orozco Camacho

Reihaneh Rabbany

In modern days, social media platforms provide accessible channels for the inter-1 action and immediate reﬂection of the most important ev… (voir plus)ents happening around 2 the world. In this paper, we, ﬁrstly, present a curated set of datasets whose origin 3 stem from the Twitter’s Information Operations 1 efforts. More notably, these 4 accounts, which have been already suspended, provide a notion of how state-backed 5 human trolls operate. 6 Secondly, we present detailed analyses of how these behaviours vary over time, 7 and motivate its use and abstraction in the context of deep representation learning: 8 for instance, to learn and, potentially track, troll behaviour. We present baselines 9 for such tasks and highlight the differences there may exist within the literature. 10 Finally, we utilize the representations learned for behaviour prediction to classify 11 trolls from "real" users, using a sample of non-suspended active accounts. 12

2021-12-06

LatinX in AI at Neural Information Processing Systems Conference 2021 (publié)

doi.org

openreview.net

Deep Reinforcement Learning at the Edge of the Statistical Precipice

Marc G. Bellemare

Deep reinforcement learning (RL) algorithms are predominantly evaluated by comparing their relative performance on a large suite of tasks. M… (voir plus)ost published results on deep RL benchmarks compare point estimates of aggregate performance such as mean and median scores across tasks, ignoring the statistical uncertainty implied by the use of a finite number of training runs. Beginning with the Arcade Learning Environment (ALE), the shift towards computationally-demanding benchmarks has led to the practice of evaluating only a small number of runs per task, exacerbating the statistical uncertainty in point estimates. In this paper, we argue that reliable evaluation in the few run deep RL regime cannot ignore the uncertainty in results without running the risk of slowing down progress in the field. We illustrate this point using a case study on the Atari 100k benchmark, where we find substantial discrepancies between conclusions drawn from point estimates alone versus a more thorough statistical analysis. With the aim of increasing the field's confidence in reported results with a handful of runs, we advocate for reporting interval estimates of aggregate performance and propose performance profiles to account for the variability in results, as well as present more robust and efficient aggregate metrics, such as interquartile mean scores, to achieve small uncertainty in results. Using such statistical tools, we scrutinize performance evaluations of existing algorithms on other widely used RL benchmarks including the ALE, Procgen, and the DeepMind Control Suite, again revealing discrepancies in prior comparisons. Our findings call for a change in how we evaluate performance in deep RL, for which we present a more rigorous evaluation methodology, accompanied with an open-source library rliable, to prevent unreliable results from stagnating the field.

2021-12-06

Neural Information Processing Systems (Accept (Oral))

doi.org

openreview.net

Discrete-Valued Neural Communication in Structured Architectures Enhances Generalization

Dianbo Liu

Alex Lamb

Kenji Kawaguchi

Anirudh Goyal

Chen Sun

Michael C. Mozer

Yoshua Bengio

Deep learning has advanced from fully connected architectures to structured models organized into components, e.g., the transformer composed… (voir plus) of positional elements, modular architectures divided into slots, and graph neural nets made up of nodes. In structured models, an interesting question is how to conduct dynamic and possibly sparse communication among the separate components. Here, we explore the hypothesis that restricting the transmitted information among components to discrete representations is a beneficial bottleneck. The motivating intuition is human language in which communication occurs through discrete symbols. Even though individuals have different understandings of what a "cat" is based on their specific experiences, the shared discrete token makes it possible for communication among individuals to be unimpeded by individual differences in internal representation. To discretize the values of concepts dynamically communicated among specialist components, we extend the quantization mechanism from the Vector-Quantized Variational Autoencoder to multi-headed discretization with shared codebooks and use it for discrete-valued neural communication (DVNC). Our experiments show that DVNC substantially improves systematic generalization in a variety of architectures -- transformers, modular architectures, and graph neural networks. We also show that the DVNC is robust to the choice of hyperparameters, making the method very useful in practice. Moreover, we establish a theoretical justification of our discretization process, proving that it has the ability to increase noise robustness and reduce the underlying dimensionality of the model.

2021-12-06

Advances in Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

Dynamic Inference with Neural Interpreters

Nasim Rahaman

Muhammad Waleed Gondal

Shruti Joshi

Peter Gehler

Yoshua Bengio

Francesco Locatello

Bernhard Schölkopf

Modern neural network architectures can leverage large amounts of data to generalize well within the training distribution. However, they ar… (voir plus)e less capable of systematic generalization to data drawn from unseen but related distributions, a feat that is hypothesized to require compositional reasoning and reuse of knowledge. In this work, we present Neural Interpreters, an architecture that factorizes inference in a self-attention network as a system of modules, which we call \emph{functions}. Inputs to the model are routed through a sequence of functions in a way that is end-to-end learned. The proposed architecture can flexibly compose computation along width and depth, and lends itself well to capacity extension after training. To demonstrate the versatility of Neural Interpreters, we evaluate it in two distinct settings: image classification and visual abstract reasoning on Raven Progressive Matrices. In the former, we show that Neural Interpreters perform on par with the vision transformer using fewer parameters, while being transferrable to a new task in a sample efficient manner. In the latter, we find that Neural Interpreters are competitive with respect to the state-of-the-art in terms of systematic generalization

2021-12-06

Conference on Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

Dynamics of Stochastic Momentum Methods on Large-scale, Quadratic Models

Courtney Paquette

Elliot Paquette

We analyze a class of stochastic gradient algorithms with momentum on a high-dimensional random least squares problem. Our framework, inspir… (voir plus)ed by random matrix theory, provides an exact (deterministic) characterization for the sequence of loss values produced by these algorithms which is expressed only in terms of the eigenvalues of the Hessian. This leads to simple expressions for nearly-optimal hyperparameters, a description of the limiting neighborhood, and average-case complexity. As a consequence, we show that (small-batch) stochastic heavy-ball momentum with a fixed momentum parameter provides no actual performance improvement over SGD when step sizes are adjusted correctly. For contrast, in the non-strongly convex setting, it is possible to get a large improvement over SGD using momentum. By introducing hyperparameters that depend on the number of samples, we propose a new algorithm sDANA (stochastic dimension adjusted Nesterov acceleration) which obtains an asymptotically optimal average-case complexity while remaining linearly convergent in the strongly convex setting without adjusting parameters.

2021-12-06

Conference on Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

End-to-End Training of Multi-Document Reader and Retriever for Open-Domain Question Answering

Devendra Singh Sachan

Siva Reddy

William Hamilton

Chris Dyer

Dani Yogatama

We present an end-to-end differentiable training method for retrieval-augmented open-domain question answering systems that combine informat… (voir plus)ion from multiple retrieved documents when generating answers. We model retrieval decisions as latent variables over sets of relevant documents. Since marginalizing over sets of retrieved documents is computationally hard, we approximate this using an expectation-maximization algorithm. We iteratively estimate the value of our latent variable (the set of relevant documents for a given question) and then use this estimate to update the retriever and reader parameters. We hypothesize that such end-to-end training allows training signals to flow to the reader and then to the retriever better than staged-wise training. This results in a retriever that is able to select more relevant documents for a question and a reader that is trained on more accurate documents to generate an answer. Experiments on three benchmark datasets demonstrate that our proposed method outperforms all existing approaches of comparable size by 2-3% absolute exact match points, achieving new state-of-the-art results. Our results also demonstrate the feasibility of learning to retrieve to improve answer generation without explicit supervision of retrieval decisions.

2021-12-06

Conference on Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

Flexible Option Learning

Martin Klissarov

Doina Precup

Temporal abstraction in reinforcement learning (RL), offers the promise of improving generalization and knowledge transfer in complex enviro… (voir plus)nments, by propagating information more efficiently over time. Although option learning was initially formulated in a way that allows updating many options simultaneously, using off-policy, intra-option learning (Sutton, Precup & Singh, 1999), many of the recent hierarchical reinforcement learning approaches only update a single option at a time: the option currently executing. We revisit and extend intra-option learning in the context of deep reinforcement learning, in order to enable updating all options consistent with current primitive action choices, without introducing any additional estimates. Our method can therefore be naturally adopted in most hierarchical RL frameworks. When we combine our approach with the option-critic algorithm for option discovery, we obtain significant improvements in performance and data-efficiency across a wide variety of domains.

2021-12-06

NEURIPS (Accept (Spotlight))

doi.org

openreview.net

Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation

This paper is about the problem of learning a stochastic policy for generating an object (like a molecular graph) from a sequence of actions… (voir plus), such that the probability of generating an object is proportional to a given positive reward for that object. Whereas standard return maximization tends to converge to a single return-maximizing sequence, there are cases where we would like to sample a diverse set of high-return solutions. These arise, for example, in black-box function optimization when few rounds are possible, each with large batches of queries, where the batches should be diverse, e.g., in the design of new molecules. One can also see this as a problem of approximately converting an energy function to a generative distribution. While MCMC methods can achieve that, they are expensive and generally only perform local exploration. Instead, training a generative policy amortizes the cost of search during training and yields to fast generation. Using insights from Temporal Difference learning, we propose GFlowNet, based on a view of the generative process as a flow network, making it possible to handle the tricky case where different trajectories can yield the same final state, e.g., there are many ways to sequentially add atoms to generate some molecular graph. We cast the set of trajectories as a flow and convert the flow consistency equations into a learning objective, akin to the casting of the Bellman equations into Temporal Difference methods. We prove that any global minimum of the proposed objectives yields a policy which samples from the desired distribution, and demonstrate the improved performance and diversity of GFlowNet on a simple domain where there are many modes to the reward function, and on a molecule synthesis task.

2021-12-06

Conference on Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

Gradient Starvation: A Learning Proclivity in Neural Networks

Sékou-Oumar Kaba

We identify and formalize a fundamental gradient descent phenomenon resulting in a learning proclivity in over-parameterized neural networks… (voir plus). Gradient Starvation arises when cross-entropy loss is minimized by capturing only a subset of features relevant for the task, despite the presence of other predictive features that fail to be discovered. This work provides a theoretical explanation for the emergence of such feature imbalance in neural networks. Using tools from Dynamical Systems theory, we identify simple properties of learning dynamics during gradient descent that lead to this imbalance, and prove that such a situation can be expected given certain statistical structure in training data. Based on our proposed formalism, we develop guarantees for a novel regularization method aimed at decoupling feature learning dynamics, improving accuracy and robustness in cases hindered by gradient starvation. We illustrate our findings with simple and real-world out-of-distribution (OOD) generalization experiments.

2021-12-06

NEURIPS (Accept (Poster))

doi.org

openreview.net

How to transfer algorithmic reasoning knowledge to learn new algorithms?

Louis-Pascal A. C. Xhonneux

Andreea Deac

Petar Veličković

Jian Tang

Learning to execute algorithms is a fundamental problem that has been widely studied. Prior work~\cite{veli19neural} has shown that to enabl… (voir plus)e systematic generalisation on graph algorithms it is critical to have access to the intermediate steps of the program/algorithm. In many reasoning tasks, where algorithmic-style reasoning is important, we only have access to the input and output examples. Thus, inspired by the success of pre-training on similar tasks or data in Natural Language Processing (NLP) and Computer Vision, we set out to study how we can transfer algorithmic reasoning knowledge. Specifically, we investigate how we can use algorithms for which we have access to the execution trace to learn to solve similar tasks for which we do not. We investigate two major classes of graph algorithms, parallel algorithms such as breadth-first search and Bellman-Ford and sequential greedy algorithms such as Prim and Dijkstra. Due to the fundamental differences between algorithmic reasoning knowledge and feature extractors such as used in Computer Vision or NLP, we hypothesise that standard transfer techniques will not be sufficient to achieve systematic generalisation. To investigate this empirically we create a dataset including 9 algorithms and 3 different graph types. We validate this empirically and show how instead multi-task learning can be used to achieve the transfer of algorithmic reasoning knowledge.

2021-12-06

Conference on Neural Information Processing Systems (Accept (Poster))

doi.org

openreview.net

Formation d'entrepreneurs scientifiques

Boussole des politiques en IA

Demandes de supervision

Publications

Formation d'entrepreneurs scientifiques

Boussole des politiques en IA

Demandes de supervision

Mots-clés populaires:

Publications