A Unified Game-Theoretic Approach to Multiagent Reinforcement Learning
–Neural Information Processing Systems
In this paper, we first observe that policies learned using InRL can overfit to the other agents' policies during training, failing to sufficiently generalize during execution. We introduce a new metric, joint-policy correlation, to quantify this effect.
Neural Information Processing Systems
Nov-21-2025, 06:53:02 GMT
- Country:
- Asia > Middle East
- Jordan (0.04)
- Europe > United Kingdom
- England > Cambridgeshire > Cambridge (0.04)
- North America
- Canada
- United States
- California
- Los Angeles County
- Long Beach (0.04)
- Los Angeles (0.14)
- San Francisco County > San Francisco (0.14)
- Los Angeles County
- New Jersey > Middlesex County
- New Brunswick (0.04)
- Texas > Travis County
- Austin (0.04)
- California
- Oceania > Australia
- New South Wales > Sydney (0.04)
- Asia > Middle East
- Genre:
- Research Report (0.46)
- Industry:
- Leisure & Entertainment > Games > Computer Games (0.93)
- Technology: