Anirudh Goyal

Neural Function Modules with Sparse Arguments: A Dynamic Approach to Integrating Information across Layers

Alex Lamb

Agnieszka Słowik

Michael Mozer

Philippe Beaudoin

Feed-forward neural networks consist of a sequence of layers, in which each layer performs some processing on the information from the previ… (voir plus)ous layer. A downside to this approach is that each layer (or module, as multiple modules can operate in parallel) is tasked with processing the entire hidden state, rather than a particular part of the state which is most relevant for that module. Methods which only operate on a small number of input variables are an essential part of most programming languages, and they allow for improved modularity and code re-usability. Our proposed method, Neural Function Modules (NFM), aims to introduce the same structural capability into deep learning. Most of the work in the context of feed-forward networks combining top-down and bottom-up feedback is limited to classification problems. The key contribution of our work is to combine attention, sparsity, top-down and bottom-up feedback, in a flexible algorithm which, as we show, improves the results in standard classification, out-of-domain generalization, generative modeling, and learning representations in the context of reinforcement learning.

2021-04-12

International Conference on Artificial Intelligence and Statistics (Poster)

proceedings.mlr.press

Transformers with Competitive Ensembles of Independent Mechanisms

Alex Lamb

Di He

Guolin Ke

Chien-Feng Liao

Mirco Ravanelli

An important development in deep learning from the earliest MLPs has been a move towards architectures with structural inductive biases whic… (voir plus)h enable the model to keep distinct sources of information and routes of processing well-separated. This structure is linked to the notion of independent mechanisms from the causality literature, in which a mechanism is able to retain the same processing as irrelevant aspects of the world are changed. For example, convnets enable separation over positions, while attention-based architectures (especially Transformers) learn which combination of positions to process dynamically. In this work we explore a way in which the Transformer architecture is deficient: it represents each position with a large monolithic hidden representation and a single set of parameters which are applied over the entire hidden representation. This potentially throws unrelated sources of information together, and limits the Transformer's ability to capture independent mechanisms. To address this, we propose Transformers with Independent Mechanisms (TIM), a new Transformer layer which divides the hidden representation and parameters into multiple mechanisms, which only exchange information through attention. Additionally, we propose a competition mechanism which encourages these mechanisms to specialize over time steps, and thus be more independent. We study TIM on a large-scale BERT model, on the Image Transformer, and on speech enhancement and find evidence for semantically meaningful specialization as well as improved performance.

2021-02-26

ArXiv (prépublication)

openreview.net

Towards Causal Representation Learning

Bernhard Schölkopf

Francesco Locatello

Stefan Bauer

Nan Rosemary Ke

Nal Kalchbrenner

The two fields of machine learning and graphical causality arose and developed separately. However, there is now cross-pollination and incre… (voir plus)asing interest in both fields to benefit from the advances of the other. In the present paper, we review fundamental concepts of causal inference and relate them to crucial open problems of machine learning, including transfer and generalization, thereby assaying how causality can contribute to modern machine learning research. This also applies in the opposite direction: we note that most work in causality starts from the premise that the causal variables are given. A central problem for AI and causality is, thus, causal representation learning, the discovery of high-level causal variables from low-level observations. Finally, we delineate some implications of causality for machine learning and propose key research areas at the intersection of both communities.

2021-02-21

ArXiv (prépublication)

Episodes Meta Sequence S 2 Fast Update Slow Update Fast Update Slow Update

Kanika Madan

Nan Rosemary Ke

Bernhard Schölkopf

Decomposing knowledge into interchangeable pieces promises a generalization advantage when there are changes in distribution. A learning age… (voir plus)nt interacting with its environment is likely to be faced with situations requiring novel combinations of existing pieces of knowledge. We hypothesize that such a decomposition of knowledge is particularly relevant for being able to generalize in a systematic manner to out-of-distribution changes. To study these ideas, we propose a particular training framework in which we assume that the pieces of knowledge an agent needs and its reward function are stationary and can be re-used across tasks. An attention mechanism dynamically selects which modules can be adapted to the current task, and the parameters of the selected modules are allowed to change quickly as the learner is confronted with variations in what it experiences, while the parameters of the attention mechanisms act as stable, slowly changing, metaparameters.We focus on pieces of knowledge captured by an ensemble of modules sparsely communicating with each other via a bottleneck of attention. We find that meta-learning the modular aspects of the proposed system greatly helps in achieving faster adaptation in a reinforcement learning setup involving navigation in a partially observed grid world with image-level input. We also find that reversing the role of parameters and meta-parameters does not work nearly as well, suggesting a particular role for fast adaptation of the dynamically selected modules.

2020-12-31

(publié)

www.semanticscholar.org

Untangling tradeoffs between recurrence and self-attention in neural networks

2020-12-06

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

Learning to Combine Top-Down and Bottom-Up Signals in Recurrent Neural Networks with Attention over Modules

Murray Shanahan

Michael Mozer

Robust perception relies on both bottom-up and top-down signals. Bottom-up signals consist of what's directly observed through sensation. To… (voir plus)p-down signals consist of beliefs and expectations based on past experience and short-term memory, such as how the phrase `peanut butter and~...' will be completed. The optimal combination of bottom-up and top-down information remains an open question, but the manner of combination must be dynamic and both context and task dependent. To effectively utilize the wealth of potential top-down information available, and to prevent the cacophony of intermixed signals in a bidirectional architecture, mechanisms are needed to restrict information flow. We explore deep recurrent neural net architectures in which bottom-up and top-down signals are dynamically combined using attention. Modularity of the architecture further restricts the sharing and communication of information. Together, attention and modularity direct information flow, which leads to reliable performance improvements in perceptual and language tasks, and in particular improves robustness to distractions and noisy data. We demonstrate on a variety of benchmarks in language modeling, sequential image classification, video prediction and reinforcement learning that the \emph{bidirectional} information flow can improve results over strong baselines.

2020-11-20

Proceedings of the 37th International Conference on Machine Learning (publié)

proceedings.mlr.press

Small-GAN: Speeding Up GAN Training Using Core-Sets

Samarth Sinha

Han Zhang

Hugo Larochelle

Augustus Odena

Recent work by Brock et al. (2018) suggests that Generative Adversarial Networks (GANs) benefit disproportionately from large mini-batch siz… (voir plus)es. Unfortunately, using large batches is slow and expensive on conventional hardware. Thus, it would be nice if we could generate batches that were effectively large though actually small. In this work, we propose a method to do this, inspired by the use of Coreset-selection in active learning. When training a GAN, we draw a large batch of samples from the prior and then compress that batch using Coreset-selection. To create effectively large batches of 'real' images, we create a cached dataset of Inception activations of each training image, randomly project them down to a smaller dimension, and then use Coreset-selection on those projected activations at training time. We conduct experiments showing that this technique substantially reduces training time and memory usage for modern GAN variants, that it reduces the fraction of dropped modes in a synthetic dataset, and that it allows GANs to reach a new state of the art in anomaly detection.

2020-11-20

Proceedings of the 37th International Conference on Machine Learning (publié)

proceedings.mlr.press

S2RMs: Spatially Structured Recurrent Modules

Nasim Rahaman

Muhammad Waleed Gondal

Manuel Wuthrich

Stefan Bauer

Y. Sharma

Bernhard Schölkopf

2020-07-12

ArXiv (prépublication)

Object Files and Schemata: Factorizing Declarative and Procedural Knowledge in Dynamical Systems

Sergey Levine

Charles Blundell

Michael Mozer

2020-06-28

arXiv (prépublication)

Learning Long-term Dependencies Using Cognitive Inductive Biases in Self-attention RNNs

Attention and self-attention mechanisms, inspired by cognitive processes, are now central to state-of-the-art deep learning on sequential ta… (voir plus)sks. However, most recent progress hinges on heuristic approaches that rely on considerable memory and computational resources that scale poorly. In this work, we propose a relevancy screening mechanism, inspired by the cognitive process of memory consolidation, that allows for a scalable use of sparse self-attention with recurrence. We use simple numerical experiments to demonstrate that this mechanism helps enable recurrent systems on generalization and transfer learning tasks. Based on our results, we propose a concrete direction of research to improve scalability and generalization of attentive recurrent networks.

2019-12-31

(publié)

www.semanticscholar.org

Learning the Arrow of Time for Problems in Reinforcement Learning

Nasim Rahaman

Steffen Wolf

Roman Remme