An Efficient Application of Neuroevolution for Competitive Multiagent Learning

TL;DR

This paper demonstrates an efficient neuroevolutionary approach using NEAT for competitive multiagent learning in a modified pong environment, achieving rapid training and optimal policies.

Contribution

It introduces a novel neuroevolutionary method that efficiently learns competitive policies for multiple agents with different rules in a shared environment.

Findings

Achieves optimal agent behavior in short training time

Outperforms existing multiagent reinforcement learning models

Utilizes environment properties to simplify the neuroevolution process

Abstract

Multiagent systems provide an ideal environment for the evaluation and analysis of real-world problems using reinforcement learning algorithms. Most traditional approaches to multiagent learning are affected by long training periods as well as high computational complexity. NEAT (NeuroEvolution of Augmenting Topologies) is a popular evolutionary strategy used to obtain the best performing neural network architecture often used to tackle optimization problems in the field of artificial intelligence. This paper utilizes the NEAT algorithm to achieve competitive multiagent learning on a modified pong game environment in an efficient manner. The competing agents abide by different rules while having similar observation space parameters. The proposed algorithm utilizes this property of the environment to define a singular neuroevolutionary procedure that obtains the optimal policy for all the agents. The compiled results indicate that the proposed implementation achieves ideal behaviour in a very short training period when compared to existing multiagent reinforcement learning models.