Coevolutionary Neural Population Models

TL;DR

This paper introduces a neural network-based approach to model evolutionary population dynamics, drawing parallels with adversarial training in deep learning, and demonstrates the applicability of evolutionary game theory to these models.

Contribution

It proposes a novel method combining neural networks with evolutionary dynamics and shows how game theory can describe neural population behaviors.

Findings

Neural networks can effectively model evolutionary population dynamics.

Adversarial training concepts relate to coevolutionary neural systems.

Evolutionary game theory describes neural population behaviors.

Abstract

We present a method for using neural networks to model evolutionary population dynamics, and draw parallels to recent deep learning advancements in which adversarially-trained neural networks engage in coevolutionary interactions. We conduct experiments which demonstrate that models from evolutionary game theory are capable of describing the behavior of these neural population systems.