Analysis of the difficulty of learning goal-scoring behaviour for robot soccer

Learning goal-scoring behaviour from scratch for simulated robot soccer is considered to be a very difficult problem, and is often achieved by endowing players with an innate set of hand-coded skills, or by decomposing the problem into learning a set of simpler behaviours which are then aggregated into goal-scoring behaviour. When only basic skills are available to the player the fitness landscape is very flat, containing only a few thin peaks. As more human expertise is injected via hand-coded skills or a composite fitness function, more gradient information becomes apparent on the landscape and the genetic search is more successful. The work presented in this paper uses autocorrelation and information content measures to examine features of the fitness landscape to explain how the difficulty of the problem is changed by injecting human expertise. Track: Learning Classifier Systems And Other Genetics-Based Machine Learning Categories and Subject Descriptors F.2.0 [Theory of Computati...

• Type: Conference
• Year: 2006
• Where: GECCO
• Authors: Jeff Riley, Victor Ciesielski
• Keywords: Fitness Landscapes, GECCO 2006, Goal-scoring Behaviour, Hand-coded Skills, Optimization