Reinforcement Learning Produces Dominant Strategies for the Iterated Prisoner's Dilemma

19 Jul 2017 · Marc Harper, Vincent Knight, Martin Jones, Georgios Koutsovoulos, Nikoleta E. Glynatsi, Owen Campbell ·

We present tournament results and several powerful strategies for the Iterated Prisoner's Dilemma created using reinforcement learning techniques (evolutionary and particle swarm algorithms). These strategies are trained to perform well against a corpus of over 170 distinct opponents, including many well-known and classic strategies. All the trained strategies win standard tournaments against the total collection of other opponents. The trained strategies and one particular human made designed strategy are the top performers in noisy tournaments also.

PDF Abstract

Code

Add Remove Mark official

Axelrod-Python/Axelrod

699

Datasets

Add Datasets introduced or used in this paper

Edit Social Preview

Reinforcement Learning Produces Dominant Strategies for the Iterated Prisoner's Dilemma

Code Edit Add Remove Mark official

Categories

Datasets Edit

Code

Add Remove Mark official

Datasets