Timezone: »

Learning to Coordinate with Coordination Graphs in Repeated Single-Stage Multi-Agent Decision Problems
Eugenio Bargiacchi · Timothy Verstraeten · Diederik Roijers · Ann Nowé · Hado van Hasselt

Thu Jul 12 09:15 AM -- 12:00 PM (PDT) @ Hall B #126

Learning to coordinate between multiple agents is an important problem in many reinforcement learning problems. Key to learning to coordinate is exploiting loose couplings, i.e., conditional independences between agents. In this paper we study learning in repeated fully cooperative games, multi-agent multi-armed bandits (MAMABs), in which the expected rewards can be expressed as a coordination graph. We propose multi-agent upper confidence exploration (MAUCE), a new algorithm for MAMABs that exploits loose couplings, which enables us to prove a regret bound that is logarithmic in the number of arm pulls and only linear in the number of agents. We empirically compare MAUCE to sparse cooperative Q-learning, and a state-of-the-art combinatorial bandit approach, and show that it performs much better on a variety of settings, including learning control policies for wind farms.

Author Information

Eugenio Bargiacchi (Vrije Universiteit Brussel)
Timothy Verstraeten (Vrije Universiteit Brussel)
Diederik Roijers (VUB)
Ann Nowé (Vrije Universiteit Brussel)
Hado van Hasselt (DeepMind)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors