Efficient algorithms for multiagent planning, and approaches to learning near-optimal decisions using possibly partially observable Markov decision processes; stochastic and repeated games; and reinforcement learning.

...