Metraipolis › archive › dc_565709a56e17

Uncertainty-Driven Self-Play for Robust Multi-Agent Driving Policies — arXiv:2610.19714

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19714
License (as archived)
CC0-1.0
Archived
2026-10-15T13:50:44.839Z
SHA-256 of the archived text
41aaddf6c236df3fc697865eaa47990584c0b5b499025a4ef3911a2e8b7efee5
Stable URL of this page
https://metraipolis.com/documento/dc_565709a56e17

How to cite

Uncertainty-Driven Self-Play for Robust Multi-Agent Driving Policies — arXiv:2610.19714 — arXiv (en), https://arxiv.org/abs/2610.19714 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_565709a56e17 (Metraipolis archive dc_565709a56e17, sha256 41aaddf6c236df3fc697865eaa47990584c0b5b499025a4ef3911a2e8b7efee5)
BibTeX
@misc{dc565709a56e17,
  title = {Uncertainty-Driven Self-Play for Robust Multi-Agent Driving Policies — arXiv:2610.19714},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19714},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_565709a56e17; sha256 41aaddf6c236df3fc697865eaa47990584c0b5b499025a4ef3911a2e8b7efee5},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19714. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Learning robust driving policies that generalize beyond their training domain remains a central challenge in motion planning. We propose an uncertainty-driven self-play scheme, in which half of the agents are rewarded for increasing their neighbors' epistemic uncertainty, while the other half learn to drive safely as the exploring agents continuously seek out their weaknesses. Epistemic uncertainty indicates where the policy is least confident and serves as an automatic proxy for scenario difficulty. Exploring agents use this signal to seek out hard cases, generating a challenging curriculum without manual scenario design. The resulting policy outperforms standard self-play, achieving best-in-class zero-shot transfer to nuplan Val14 (99.12% vs. 97.97%) and Argoverse 2 (82.39% vs. 78.03%) without fine-tuning, while also improving performance over standard self-play on Waymo. Our diversity-driven approach achieves 85% higher LSTM hidden-state coverage than the baseline, indicating a much richer set of behaviors. Further analysis shows that epistemic uncertainty serves as a calibrated difficulty detector, with highest-uncertain agents exhibiting 14.8 times the combined failure rate across collision, off-road, and DNF, of lowest-uncertain agents, enabling automatic prioritization of challenging scenarios without ground-truth labels.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.19714 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T13:48:35.458Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).