Stackelberg Games
Overview
Stackelberg games model sequential strategic decisions: a leader chooses a policy or commitment while anticipating how followers will respond. Researchers analyze equilibria and develop learning methods including regret minimization, calibration, bilevel optimization, and reinforcement learning, with applications in security, markets, networks, and multi-agent control. Work also studies how to learn when follower behavior or rewards are unknown, using structured complexity bounds, active exploration, and benchmarks for adaptive opponents. A central challenge is making guarantees computationally tractable and reliable when responses are uncertain, changing, or more complex than the assumed model.
Latest papers
20 papers1 paper in the last four weeks, down 75% on the four weeks before. 0.0% of all new papers.