Aaron Courville

Rishabh Agarwal

Deep reinforcement learning (RL) agents trained on a few environments, often struggle to generalize on unseen environments, even when such e… (voir plus)nvironments are semantically equivalent to training environments. Such agents learn representations that overfit the characteristics of the training environments. We posit that generalization can be improved by assigning similar representations to scenarios with similar sequences of long-term optimal behavior. To do so, we propose behavior predictive representations (BPR) that capture long-term optimal behavior. BPR trains an agent to predict latent state representations multiple steps into the future such that these representations can predict the optimal behavior at the future steps. We demonstrate that BPR provides large gains on a jumping task from pixels, a problem designed to test generalization.

2021-12-12

NeurIPS.cc/2021/Workshop/DeepRL (accepté)

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.

2021-11-08

NeurIPS.cc/2021/Conference (poster)

openreview.net

Generative Compositional Augmentations for Scene Graph Prediction

Boris Knyazev

Harm de Vries

Cătălina Cangea

Graham W. Taylor

Eugene Belilovsky

Inferring objects and their relationships from an image in the form of a scene graph is useful in many applications at the intersection of v… (voir plus)ision and language. We consider a challenging problem of compositional generalization that emerges in this task due to a long tail data distribution. Current scene graph generation models are trained on a tiny fraction of the distribution corresponding to the most frequent compositions, e.g. . However, test images might contain zero- and few-shot compositions of objects and relationships, e.g. . Despite each of the object categories and the predicate (e.g. 'on') being frequent in the training data, the models often fail to properly understand such unseen or rare compositions. To improve generalization, it is natural to attempt increasing the diversity of the training distribution. However, in the graph domain this is non-trivial. To that end, we propose a method to synthesize rare yet plausible scene graphs by perturbing real ones. We then propose and empirically study a model based on conditional generative adversarial networks (GANs) that allows us to generate visual features of perturbed scene graphs and learn from them in a joint fashion. When evaluated on the Visual Genome dataset, our approach yields marginal, but consistent improvements in zero- and few-shot metrics. We analyze the limitations of our approach indicating promising directions for future research.

2021-09-30

2021 IEEE/CVF International Conference on Computer Vision (ICCV) (publié)

Haptics-based Curiosity for Sparse-Reward Tasks

Sai Rajeswar

Cyril Ibrahim

Nitin Surya

Florian Golemo

David Vázquez

Pedro O. Pinheiro

2021-09-12

robot-learning.org/CoRL/2021/Conference (poster)

proceedings.mlr.press

StructFormer: Joint Unsupervised Induction of Dependency and Constituency Structure from Masked Language Modeling

Yikang Shen

Yi Tay

Che Zheng

Dara Bahri

Donald Metzler

There are two major classes of natural language grammar -- the dependency grammar that models one-to-one correspondences between words and t… (voir plus)he constituency grammar that models the assembly of one or several corresponded words. While previous unsupervised parsing methods mostly focus on only inducing one class of grammars, we introduce a novel model, StructFormer, that can simultaneously induce dependency and constituency structure. To achieve this, we propose a new parsing framework that can jointly generate a constituency tree and dependency graph. Then we integrate the induced dependency relations into the transformer, in a differentiable manner, through a novel dependency-constrained self-attention mechanism. Experimental results show that our model can achieve strong results on unsupervised constituency parsing, unsupervised dependency parsing, and masked language modeling at the same time.

2021-07-31

Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) (publié)

openreview.net

Continuous Coordination As a Realistic Scenario for Lifelong Learning

Hadi Nekoei

Akilesh Badrinaaraayanan

Sarath Chandar

Current deep reinforcement learning (RL) algorithms are still highly task-specific and lack the ability to generalize to new environments. L… (voir plus)ifelong learning (LLL), however, aims at solving multiple tasks sequentially by efficiently transferring and using knowledge between tasks. Despite a surge of interest in lifelong RL in recent years, the lack of a realistic testbed makes robust evaluation of LLL algorithms difficult. Multi-agent RL (MARL), on the other hand, can be seen as a natural scenario for lifelong RL due to its inherent non-stationarity, since the agents' policies change over time. In this work, we introduce a multi-agent lifelong learning testbed that supports both zero-shot and few-shot settings. Our setup is based on Hanabi -- a partially-observable, fully cooperative multi-agent game that has been shown to be challenging for zero-shot coordination. Its large strategy space makes it a desirable environment for lifelong RL tasks. We evaluate several recent MARL methods, and benchmark state-of-the-art LLL algorithms in limited memory and computation regimes to shed light on their strengths and weaknesses. This continual learning paradigm also provides us with a pragmatic way of going beyond centralized training which is the most commonly used training protocol in MARL. We empirically show that the agents trained in our setup are able to coordinate well with unseen agents, without any additional assumptions made by previous works. The code and all pre-trained models are available at https://github.com/chandar-lab/Lifelong-Hanabi.

2021-06-30

Proceedings of the 38th International Conference on Machine Learning (publié)

proceedings.mlr.press

Out-of-Distribution Generalization via Risk Extrapolation

David Krueger

Ethan Caballero

Joern-Henrik Jacobsen

Rémi Le Priol

Distributional shift is one of the major obstacles when transferring machine learning prediction systems from the lab to the real world. To … (voir plus)tackle this problem, we assume that variation across training domains is representative of the variation we might encounter at test time, but also that shifts at test time may be more extreme in magnitude. In particular, we show that reducing differences in risk across training domains can reduce a model's sensitivity to a wide range of extreme distributional shifts, including the challenging setting where the input contains both causal and anti-causal elements. We motivate this approach, Risk Extrapolation (REx), as a form of robust optimization over a perturbation set of extrapolated domains (MM-REx), and propose a penalty on the variance of training risks (V-REx) as a simpler variant. We prove that variants of REx can recover the causal mechanisms of the targets, while also providing some robustness to changes in the input distribution ("covariate shift"). By appropriately trading-off robustness to causally induced distributional shifts and covariate shift, REx is able to outperform alternative methods such as Invariant Risk Minimization in situations where these types of shift co-occur.

2021-06-30

Proceedings of the 38th International Conference on Machine Learning (publié)

proceedings.mlr.press

Hierarchical Video Generation for Complex Data

Lluis Castrejon

Nicolas Ballas

2021-06-03

ArXiv (prépublication)

Explicitly Modeling Syntax in Language Models with Incremental Parsing and a Dynamic Oracle

Syntax is fundamental to our thinking about language. Failing to capture the structure of input language could lead to generalization proble… (voir plus)ms and over-parametrization. In the present work, we propose a new syntax-aware language model: Syntactic Ordered Memory (SOM). The model explicitly models the structure with an incremental parser and maintains the conditional probability setting of a standard language model (left-to-right). To train the incremental parser and avoid exposure bias, we also propose a novel dynamic oracle, so that SOM is more robust to wrong parsing decisions. Experiments show that SOM can achieve strong results in language modeling, incremental parsing and syntactic generalization tests, while using fewer parameters than other models.

2021-05-31

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (publié)

Understanding by Understanding Not: Modeling Negation in Language Models

R Devon Hjelm

Negation is a core construction in natural language. Despite being very successful on many tasks, state-of-the-art pre-trained language mode… (voir plus)ls often handle negation incorrectly. To improve language models in this regard, we propose to augment the language modeling objective with an unlikelihood objective that is based on negated generic sentences from a raw text corpus. By training BERT with the resulting combined objective we reduce the mean top~1 error rate to 4% on the negated LAMA dataset. We also see some improvements on the negated NLI benchmarks.

2021-05-31

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (publié)

Touch-based Curiosity for Sparse-Reward Tasks

Sai Rajeswar

Cyril Ibrahim

Nitin Surya

Florian Golemo

David Vázquez