Provably Efficient Fictitious Play Policy Optimization for Zero-Sum Markov Games with Structured Transitions

Open in new window