Research Area:  Machine Learning
Traditionally, Deep Artificial Neural Networks (DNN-s) are trained through gradient descent. Recent research shows that Deep Neuroevolution (DNE) is also capable of evolving multi-million-parameter DNN-s, which proved to be particularly useful in the field of Reinforcement Learning (RL). This is mainly due to its excellent scalability and simplicity compared to the traditional MDP-based RL methods. So far, DNE has only been applied to complex single-agent problems. As evolutionary methods are a natural choice for multi-agent problems, the question arises whether DNE can also be applied in a complex multi-agent setting. In this paper, we describe and validate a new approach based on coevolution. To validate our approach, we benchmark two Deep coevolutionary Algorithms on a range of multi-agent Atari games and compare our results against the results of Ape-X DQN. Our results show that these Deep coevolutionary algorithms (1) can be successfully trained to play various games, (2) outperform Ape-X DQN in some of them, and therefore (3) show that coevolution can be a viable approach to solving complex multi-agent decision-making problems.
Keywords:  
Multi-agent reinforcement learning
Deep Neuroevolution (DNE)
Deep Artificial Neural Networks
coevolutionary Algorithms
Author(s) Name:  Daan Klijn , A. E. Eiben
Journal name:  
Conferrence name:  GECCO -21: Proceedings of the Genetic and Evolutionary Computation Conference Companion
Publisher name:  ACM
DOI:  10.1145/3449726.3459576
Volume Information:  
Paper Link:   https://dl.acm.org/doi/abs/10.1145/3449726.3459576