Learning Dynamic Mechanisms in Unknown Environments: A Reinforcement Learning Approach

We study the problem of learning optimal dynamic mechanisms in unknown environments using reinforcement learning. We propose algorithms with provable sample complexity and regret guarantees for learning revenue-maximizing mechanisms when the buyer’s valuation distribution is unknown and must be learned through interaction.

Shuang Qiu*, Boxiang Lyu*, Qinglin Meng*, Zhaoran Wang, Zhuoran Yang, Michael I. Jordan
Download Paper