Computational Intelligence for Games


Abstract

Computational Intelligence in Games is a field of research that allows to test different types of learning, search and decision making algorithms in affordable, competitive, dynamic and reproducible environments.Ā  Many different approaches from diverse families of algorithms, such as Reinforcement Learning or Evolutionary Computation, are commonly used to tackle this kind of problems, populating conference papers, journals and competitions.

In thisĀ tutorial, we will give an overview on the field of Computational Intelligence in Games, and study how game agents can be created for the 2-player trackĀ of the GVGAI Competition (www.gvgai.net).Ā The General Video Game AI framework and competition pose the problem of creating artificial intelligence that can play a wide, and in principle unlimited, range of games. Concretely, it tackles the problem of devising an algorithm that is able to play any game it is given, even if the game is not known a priori. This area of study can be seen as an approximation of General Artificial Intelligence, with very little room for game-dependent heuristics.

Theoretical part

  • Overview of the Computational Intelligence in Games field
  • Algorithms to address decision making in highly dynamic environments and 2-player games
  • Description of the General Video Game AI Framework and Competition, with special emphasis on the 2-player track.

Lab

  • Implementation of bots for the GVGAI Framework
  • Competition between the developed bots and declaration of a winner