method · first seen 14 Oct 2026
self-play for driving policies
Training multi-agent driving policies by pitting agents against each other, including agents rewarded for finding weaknesses.
Mentions per week
PapersCodeDiscussion
Earliest sources
- Uncertainty-Driven Self-Play for Robust Multi-Agent Driving PoliciesPapers · 14 Oct 2026 · Sajad Marvi, Julian Schmidt, Alessandro Canevaro, Julian Jordan, Abhinav Valada
Latest sources
- Uncertainty-Driven Self-Play for Robust Multi-Agent Driving PoliciesPapers · 14 Oct 2026 · Sajad Marvi, Julian Schmidt, Alessandro Canevaro, Julian Jordan, Abhinav Valada