A Recurrent Neural Network for Game Theoretic Decision Making

Sudeep BhatiaUniversity of Warwick, Coventry, UK
Russell GolmanCarnegie Mellon University

Abstract

We describe the properties of a connectionist network that is able to make decisions in strategic games. We use the structure of Bidirectional Associative Memory (BAM), a minimal two-layer recurrent neural network with binary activation functions and binary connection weights. We apply BAM to finite-strategy two-player games, and show that network activation in the long run is restricted to the set of rationalizable strategies. The network is not guaranteed to reach a stable activation state, but any pure strategy profile that constitutes a stable state in the network must be a pure strategy Nash equilibrium.

Files

A Recurrent Neural Network for Game Theoretic Decision Making (488 KB)



Back to Table of Contents