Log In Sign Up

IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures

by   Lasse Espeholt, et al.

In this work we aim to solve a large collection of tasks using a single reinforcement learning agent with a single set of parameters. A key challenge is to handle the increased amount of data and extended training time, which is already a problem in single task learning. We have developed a new distributed agent IMPALA (Importance-Weighted Actor Learner Architecture) that can scale to thousands of machines and achieve a throughput rate of 250,000 frames per second. We achieve stable learning at high throughput by combining decoupled acting and learning with a novel off-policy correction method called V-trace, which was critical for achieving learning stability. We demonstrate the effectiveness of IMPALA for multi-task reinforcement learning on DMLab-30 (a set of 30 tasks from the DeepMind Lab environment (Beattie et al., 2016)) and Atari-57 (all available Atari games in Arcade Learning Environment (Bellemare et al., 2013a)). Our results show that IMPALA is able to achieve better performance than previous agents, use less data and crucially exhibits positive transfer between tasks as a result of its multi-task approach.


Multi-task Deep Reinforcement Learning with PopArt

The reinforcement learning community has made great strides in designing...

Distributed Deep Reinforcement Learning: An Overview

Deep reinforcement learning (DRL) is a very active research area. Howeve...

Crossmodal Attentive Skill Learner

This paper presents the Crossmodal Attentive Skill Learner (CASL), integ...

Off-Policy Correction For Multi-Agent Reinforcement Learning

Multi-agent reinforcement learning (MARL) provides a framework for probl...

Modular Networks Prevent Catastrophic Interference in Model-Based Multi-Task Reinforcement Learning

In a multi-task reinforcement learning setting, the learner commonly ben...

Testbeds for Reinforcement Learning

We present three problems modeled after animal learning experiments desi...

Sample Factory: Egocentric 3D Control from Pixels at 100000 FPS with Asynchronous Reinforcement Learning

Increasing the scale of reinforcement learning experiments has allowed r...

Code Repositories


in progress (I'll revisit this after NIPS deadline)

view repo


A Test-Implementation of the IMPALA algorithm (by deepmind 2018)

view repo