Learning And Adaption In Multi-agent Systems