Adaption of XCS to Multi-Learner Predator/Prey Scenarios
Published: 2010 Juli
Buchtitel: Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2010)
Verlag: ACM SIGEVO
Learning classifier systems (LCSs) are rule-based evolutionary reinforcement learning systems. Today, especially variants of Wilson’s extended classifier system (XCS) are widely applied for machine learning. Despite their widespread application, LCSs have drawbacks, e. g., in multi-learner scenarios, since the Markov property is not fulfilled.
In this paper, LCSs are investigated in an instance of the generic homogeneous and non-communicating predator/prey scenario. A group of predators collaboratively observe a (randomly) moving prey as long as possible, where each predator is equipped with a single, independent XCS. Results show that improvements in learning are achieved by cleverly adapting a multi-step approach to the characteristics of the investigated scenario. Firstly, the environmental reward function is expanded to include sensory information. Secondly, the learners are equipped with a memory to store and analyze the history of local actions and given payoffs.