MCTS/EA hybrid GVGAI players and game difficulty estimation
Authors:
- 1 September 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
Abstract
In the General Video Game Playing competitions of the last years, Monte-Carlo tree search as well as Evolutionary Algorithm based controllers have been successful. However, both approaches have certain weaknesses, suggesting that certain hybrids could outperform both. We envision and experimentally compare several types of hybrids of two basic approaches, as well as some possible extensions. In order to achieve a better understanding of the games in the competition and the strength and weaknesses of different controllers, we also propose and apply a novel game difficulty estimation scheme based on several observable game characteristics.Keywords
This publication has 9 references indexed in Scilit:
- Online Evolution for Multi-action Adversarial Games2016
- Fast Evolutionary Adaptation for Monte Carlo Tree Search2014
- Knowledge-based fast evolutionary MCTS for general video game playing2014
- A video game description language for model-based or interactive learning2013
- Rolling horizon evolution versus tree search for navigation in single-player real-time games2013
- A Survey of Monte Carlo Tree Search MethodsIEEE Transactions on Computational Intelligence and AI in Games, 2012
- The Microbial Genetic Algorithm2011
- CadiaPlayer: A Simulation-Based General Game PlayerIEEE Transactions on Computational Intelligence and AI in Games, 2009
- Bandit Based Monte-Carlo Planning2006