DDA3C: Cooperative Distributed Deep Reinforcement Learning in A Group-Agent System

02/10/2022
by   Kaiyue Wu, et al.
0

It can largely benefit the reinforcement learning process of each agent if multiple agents perform their separate reinforcement learning tasks cooperatively. These tasks can be not exactly the same but still benefit from the communication behaviour between agents due to task similarities. In fact, this learning scenario is not well understood yet and not well formulated. As the first effort, we provide a detailed discussion of this scenario, and propose group-agent reinforcement learning as a formulation of the reinforcement learning problem under this scenario and a third type of reinforcement learning problem with respect to single-agent and multi-agent reinforcement learning. We propose that it can be solved with the help of modern deep reinforcement learning techniques and provide a distributed deep reinforcement learning algorithm called DDA3C (Decentralised Distributed Asynchronous Advantage Actor-Critic) that is the first framework designed for group-agent reinforcement learning. We show through experiments in the CartPole-v0 game environment that DDA3C achieved desirable performance and has good scalability.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/22/2019

Agent Modeling as Auxiliary Task for Deep Reinforcement Learning

In this paper we explore how actor-critic methods in deep reinforcement ...
research
02/27/2020

A Visual Communication Map for Multi-Agent Deep Reinforcement Learning

Multi-agent learning distinctly poses significant challenges in the effo...
research
05/17/2021

Using Distributed Reinforcement Learning for Resource Orchestration in a Network Slicing Scenario

The Network Slicing (NS) paradigm enables the partition of physical and ...
research
01/17/2022

Summarising and Comparing Agent Dynamics with Contrastive Spatiotemporal Abstraction

We introduce a data-driven, model-agnostic technique for generating a hu...
research
02/23/2022

Using Deep Reinforcement Learning with Automatic Curriculum earning for Mapless Navigation in Intralogistics

We propose a deep reinforcement learning approach for solving a mapless ...
research
05/18/2023

Sharing Lifelong Reinforcement Learning Knowledge via Modulating Masks

Lifelong learning agents aim to learn multiple tasks sequentially over a...
research
08/18/2020

Ubiquitous Distributed Deep Reinforcement Learning at the Edge: Analyzing Byzantine Agents in Discrete Action Spaces

The integration of edge computing in next-generation mobile networks is ...

Please sign up or login with your details

Forgot password? Click here to reset