The chapter provides an extensive survey of cooperative multi-agent learning, a field that combines distributed systems and artificial intelligence to solve complex problems collaboratively. The authors define cooperative multi-agent learning as the application of machine learning to problems involving multiple agents working together to solve tasks or maximize utility. They categorize the literature into two main approaches: team learning, which uses a single learner to discover joint solutions, and concurrent learning, which employs multiple simultaneous learners, often one per agent. The chapter discusses the challenges and advantages of each approach, including the large state space in team learning and the non-stationary environment in concurrent learning. It also explores different machine learning methods such as reinforcement learning and evolutionary computation, and their applications in multi-agent systems. The authors highlight the importance of credit assignment, the dynamics of learning, and the need for adaptive dynamics and problem decomposition in multi-agent learning. The survey concludes with a presentation of problem domains and resources in multi-agent learning, emphasizing the ongoing research and potential for further development in this area.The chapter provides an extensive survey of cooperative multi-agent learning, a field that combines distributed systems and artificial intelligence to solve complex problems collaboratively. The authors define cooperative multi-agent learning as the application of machine learning to problems involving multiple agents working together to solve tasks or maximize utility. They categorize the literature into two main approaches: team learning, which uses a single learner to discover joint solutions, and concurrent learning, which employs multiple simultaneous learners, often one per agent. The chapter discusses the challenges and advantages of each approach, including the large state space in team learning and the non-stationary environment in concurrent learning. It also explores different machine learning methods such as reinforcement learning and evolutionary computation, and their applications in multi-agent systems. The authors highlight the importance of credit assignment, the dynamics of learning, and the need for adaptive dynamics and problem decomposition in multi-agent learning. The survey concludes with a presentation of problem domains and resources in multi-agent learning, emphasizing the ongoing research and potential for further development in this area.