Computational Intelligence for Games


** NEW ** You may now book your place through Proficio!

 

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.  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 this tutorial, we will give an overview on the field of Computational Intelligence in Games, and study how game agents can be created for a particular 2-player game: PlanetWars, a game that featured in the Google AI Challenge in 2010, and that received more than 4000 submissions. We will analyze the latest efforts in this game, using deep neural networks and other alternative approaches.

The program will be as follows:

Theoretical part:

  • Overview of the Computational Intelligence in Games field
  • Algorithms to address decision making in highly dynamic environments
  • Description of the Planet Wars game

Lab:

  • Implementation of bots that play the game
  • Competition between the developed bots and declaration of a winner

The tutorial will be given by Prof. Simon Lucas and Dr. Diego Pérez-Liébana.