Model Adaptation · Data store

Opponent Policy League

Data storeModel AdaptationModelsarc:OpponentPolicyLeague

A store of a diverse population of agent policies, including past versions and specially trained exploiter policies, used as opponents in competitive multi-agent training.

Responsibility. Holds diverse opponent policies for self-play and league training.

Also known as: Policy population, League

Direct neighbourhood (hover for relationship types)

Relationships

is read by dependency

Multi-Agent Policy Learner abstract Ch5.12

is written by dependency

Multi-Agent Policy Learner abstract Ch5.12

Classification

Patterns: League trainingSelf-play
Risks mitigated: Rock-paper-scissors cycling in self-play

Sources

Ch5.12: T. Nguyen, "Learning-Based Decision Making Fundamentals," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 5.12. ISBN: 9798244538229.