Sequential Information Design: Markov Persuasion Process and Its Efficient Reinforcement Learning
13Zhuoran Yang (UC Berkeley)
https://simons.berkeley.edu/talks/sequential-information-design-markov-persuasion-process-and-its-efficient-reinforcement
Multi-Agent Reinforcement Learning and Bandit Learning
In today's economy, it becomes important for Internet platforms to consider the sequential information design problem to align its long-term interest with the incentives of the gig service providers. In this talk, I will introduce a novel model of sequential information design, namely the Markov persuasion processes (MPPs), where a sender, with informational advantage, seeks to persuade a stream of myopic receivers to take actions that maximize the sender's cumulative utilities in a finite horizon Markovian environment with varying prior and utility functions. Planning in MPPs thus faces the unique challenge i finding a signaling policy that is simultaneously persuasive to the myopic receivers and inducing the optimal long-term cumulative utilities of the sender. Under the online setting where the model is unknown, I will introduce a provably efficient reinforcement learning algorithm, the Optimism-Pessimism Principle for Persuasion Process (OP4), which features a novel combination of both optimism and pessimism principles and enjoys a sublinear regret upper bound.