Diff-DAC: Distributed Actor-Critic for Average Multitask Deep Reinforcement Learning

Abstract

We propose a fully distributed actor-critic algorithm approximated by deep neural networks, named \textit{Diff-DAC}, with application to single-task and to average multitask reinforcement learning (MRL). Each agent has access to data from its local task only, but it aims to learn a policy that performs well on average for the whole set of tasks. During the learning process, agents communicate their value-policy parameters to their neighbors, diffusing the information across the network, so that they converge to a common policy, with no need for a central node. The method is scalable, since the computational and communication costs per agent grow with its number of neighbors. We derive Diff-DAC's from duality theory and provide novel insights into the standard actor-critic framework, showing that it is actually an instance of the dual ascent method that approximates the solution of a linear program. Experiments suggest that Diff-DAC can outperform the single previous distributed MRL approach (i.e., Dist-MTLPS) and even the centralized architecture.

References in corpus (4)

• How transferable are features in deep neural networks?
• Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation
• Distral: Robust Multitask Reinforcement Learning
• Connecting Generative Adversarial Networks and Actor-Critic Methods

Cited by in corpus (5)

• Fully Decentralized Multi-Agent Reinforcement Learning with Networked Agents
• Multi-Agent Reinforcement Learning via Double Averaging Primal-Dual Optimization
• RODE: Learning Roles to Decompose Multi-Agent Tasks
• Off-Policy Multi-Agent Decomposed Policy Gradients
• How to Organize your Deep Reinforcement Learning Agents: The Importance of Communication Topology