ConceptAI & ML
Low-exploitable policies
Policies that minimize maximal regret among the agents and require exact matching of state-action occupancy measures.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Imitation learning1Nash gap1
Lecturers
1Connections
2The lectures that define it or assume it, and the concepts next to it.