[KongchangAI]
Concept极大极小博弈

Max-min game

A max-min game is a robust optimization framework in which one party maximizes its payoff while the other minimizes the opponent's payoff. The equilibrium outcome is the optimal guarantee under the worst case, making it applicable to adversarial settings such as deception mitigation.

Timeline (last 90 days)

Nov 10

The researchers formalize AI agent deception mitigation as an evaluation-side partially observable extensive-form game (e-PoG), a max-min framework in which the agent maximizes reward and the monitor exposes deception from complete interaction trajectories to minimize agent payoff, with a fixed verifier validating the monitor's claims and converting them into penalties.

Unverified50%

All Facts (1)

Source Articles