Machine Learning and Inductive Logic Programming for Multi-agent Systems (original) (raw)

This paper explores the intersection of Machine Learning (ML) and Multi-Agent Systems (MAS), emphasizing the significance of learning in heterogeneous environments where agents interact and adapt. It discusses the complexities of designing agents capable of learning about cooperation and competition, and presents Inductive Logic Programming (ILP) as a valuable technique for leveraging domain knowledge in agent learning. Key issues such as agent awareness, communication, and distributed learning are examined, along with open research questions that highlight future directions in this domain.