Ahmed Touati

TDprop: Does Adaptive Optimization With Jacobi Preconditioning Help Temporal Difference Learning?

We investigate whether Jacobi preconditioning, accounting for the bootstrap term in temporal difference (TD) learning, can help boost perfor… (voir plus)mance of adaptive optimizers. Our method, TDprop, computes a per-parameter learning rate based on the diagonal preconditioning of the TD update rule. We show how this can be used in both n-step returns and TD(λ). Our theoretical findings demonstrate that including this additional preconditioning information is comparable to normal semi-gradient TD if the optimal learning rate is found for both via a hyperparameter search. This matches our experimental results. In Deep RL experiments using Expected SARSA, TDprop meets or exceeds the performance of Adam in all tested games under near-optimal learning rates, but a well-tuned SGD can yield similar performance in most settings. Our findings suggest that Jacobi preconditioning may improve upon Adam in Deep RL, but despite incorporating additional information from the TD bootstrap term, may not always be better than SGD. Moreover, they suggest that more theoretical investigations are needed to understand adaptive optimizers under optimal hyperparameter regimes in TD learning: simpler methods may, surprisingly, be theoretically comparable after a hyperparameter search.

2021-05-02

Adaptive Agents and Multi-Agent Systems (publié)

doi.org

Randomized Value Functions via Multiplicative Normalizing Flows

Randomized value functions offer a promising approach towards the challenge of efficient exploration in complex environments with high dimen… (voir plus)sional state and action spaces. Unlike traditional point estimate methods, randomized value functions maintain a posterior distribution over action-space values. This prevents the agent's behavior policy from prematurely exploiting early estimates and falling into local optima. In this work, we leverage recent advances in variational Bayesian neural networks and combine these with traditional Deep Q-Networks (DQN) and Deep Deterministic Policy Gradient (DDPG) to achieve randomized value functions for high-dimensional domains. In particular, we augment DQN and DDPG with multiplicative normalizing flows in order to track a rich approximate posterior distribution over the parameters of the value function. This allows the agent to perform approximate Thompson sampling in a computationally efficient manner via stochastic gradient methods. We demonstrate the benefits of our approach through an empirical comparison in high dimensional environments.

2020-08-05

Conference on Uncertainty in Artificial Intelligence (publié)

doi.org

proceedings.mlr.press

SVRG for Policy Evaluation with Fewer Gradient Evaluations

Stochastic variance-reduced gradient (SVRG) is an optimization method originally designed for tackling machine learning problems with a fini… (voir plus)te sum structure. SVRG was later shown to work for policy evaluation, a problem in reinforcement learning in which one aims to estimate the value function of a given policy. SVRG makes use of gradient estimates at two scales. At the slower scale, SVRG computes a full gradient over the whole dataset, which could lead to prohibitive computation costs. In this work, we show that two variants of SVRG for policy evaluation could significantly diminish the number of gradient calculations while preserving a linear convergence speed. More importantly, our theoretical result implies that one does not need to use the entire dataset in every epoch of SVRG when it is applied to policy evaluation with linear function approximation. Our experiments demonstrate large computational savings provided by the proposed methods.

2020-07-10

Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence (publié)

doi.org

arxiv.org

Stochastic Neural Network with Kronecker Flow

Chin-wei Huang

Ahmed Touati

Pascal Vincent

Gintare Karolina Dziugaite

Alexandre Lacoste

Aaron Courville

Recent advances in variational inference enable the modelling of highly structured joint distributions, but are limited in their capacity to… (voir plus) scale to the high-dimensional setting of stochastic neural networks. This limitation motivates a need for scalable parameterizations of the noise generation process, in a manner that adequately captures the dependencies among the various parameters. In this work, we address this need and present the Kronecker Flow, a generalization of the Kronecker product to invertible mappings designed for stochastic neural networks. We apply our method to variational Bayesian neural networks on predictive tasks, PAC-Bayes generalization bound estimation, and approximate Thompson sampling in contextual bandits. In all setups, our methods prove to be competitive with existing methods and better than the baselines.

2020-06-02

Artificial Intelligence and Statistics (publié)

doi.org

proceedings.mlr.press

TDprop: Does Jacobi Preconditioning Help Temporal Difference Learning?

We investigate whether Jacobi preconditioning, accounting for the bootstrap term in temporal difference (TD) learning, can help boost perfor… (voir plus)mance of adaptive optimizers. Our method, TDprop, computes a per parameter learning rate based on the diagonal preconditioning of the TD update rule. We show how this can be used in both

2019-12-31

arXiv (prépublication)

doi.org

arxiv.org

Learning Dynamics Model in Reinforcement Learning by Incorporating the Long Term Future

Nan Rosemary Ke

Amanpreet Singh

Ahmed Touati

Anirudh Goyal

Yoshua Bengio

Devi Parikh

Dhruv Batra

In model-based reinforcement learning, the agent interleaves between model learning and planning. These two components are inextricably inte… (voir plus)rtwined. If the model is not able to provide sensible long-term prediction, the executed planner would exploit model flaws, which can yield catastrophic failures. This paper focuses on building a model that reasons about the long-term future and demonstrates how to use this for efficient planning and exploration. To this end, we build a latent-variable autoregressive model by leveraging recent ideas in variational inference. We argue that forcing latent variables to carry future information through an auxiliary task substantially improves long-term predictions. Moreover, by planning in the latent space, the planner's solution is ensured to be within regions where the model is valid. An exploration strategy can be devised by searching for unlikely trajectories under the model. Our method achieves higher reward faster compared to baselines on a variety of tasks and environments in both the imitation learning and model-based reinforcement learning settings.

2019-03-04

arXiv (prépublication)

doi.org

arxiv.org

Separating value functions across time-scales

Emma Brunskill

Yann Ollivier

In many finite horizon episodic reinforcement learning (RL) settings, it is desirable to optimize for the undiscounted return - in settings … (voir plus)like Atari, for instance, the goal is to collect the most points while staying alive in the long run. Yet, it may be difficult (or even intractable) mathematically to learn with this target. As such, temporal discounting is often applied to optimize over a shorter effective planning horizon. This comes at the risk of potentially biasing the optimization target away from the undiscounted goal. In settings where this bias is unacceptable - where the system must optimize for longer horizons at higher discounts - the target of the value function approximator may increase in variance leading to difficulties in learning. We present an extension of temporal difference (TD) learning, which we call TD(

2019-02-04

ArXiv (prépublication)

doi.org

proceedings.mlr.press

Modeling the Long Term Future in Model-Based Reinforcement Learning

Nan Rosemary Ke

Amanpreet Singh

Ahmed Touati

Anirudh Goyal

Yoshua Bengio

Devi Parikh

Dhruv Batra

2018-12-31

International Conference on Learning Representations (poster)

openreview.net

Convergent Tree Backup and Retrace with Function Approximation

Off-policy learning is key to scaling up reinforcement learning as it allows to learn about a target policy from the experience generated by… (voir plus) a different behavior policy. Unfortunately, it has been challenging to combine off-policy learning with function approximation and multi-step bootstrapping in a way that leads to both stable and efficient algorithms. In this work, we show that the \textsc{Tree Backup} and \textsc{Retrace} algorithms are unstable with linear function approximation, both in theory and in practice with specific examples. Based on our analysis, we then derive stable and efficient gradient-based algorithms using a quadratic convex-concave saddle-point formulation. By exploiting the problem structure proper to these algorithms, we are able to provide convergence guarantees and finite-sample bounds. The applicability of our new analysis also goes beyond \textsc{Tree Backup} and \textsc{Retrace} and allows us to provide new convergence rates for the GTD and GTD2 algorithms without having recourse to projections or Polyak averaging.

2018-07-02

Proceedings of the 35th International Conference on Machine Learning (publié)

doi.org

proceedings.mlr.press

Learnable Explicit Density for Continuous Latent Space and Variational Inference

In this paper, we study two aspects of the variational autoencoder (VAE): the prior distribution over the latent variables and its correspon… (voir plus)ding posterior. First, we decompose the learning of VAEs into layerwise density estimation, and argue that having a flexible prior is beneficial to both sample generation and inference. Second, we analyze the family of inverse autoregressive flows (inverse AF) and show that with further improvement, inverse AF could be used as universal approximation to any complicated posterior. Our analysis results in a unified approach to parameterizing a VAE, without the need to restrict ourselves to use factorial Gaussians in the latent real space.

2017-10-05

arXiv (prépublication)

doi.org

arxiv.org

La plateforme Mila Ventures

Boussole des politiques en IA

Séances d'information virtuelles

Ahmed Touati

Publications

La plateforme Mila Ventures

Boussole des politiques en IA

Séances d'information virtuelles

Mots-clés populaires:

Ahmed Touati

Publications