Metraipolis › archive › dc_2a2a34e2c483
Beyond Expected Values: Risk-Sensitive Planning with Distributional Monte-Carlo Tree Search — arXiv:2610.23565
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.23565
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T09:10:18.831Z
- SHA-256 of the archived text
1683d54ac7d7314c4b347a5fd55a49ae6f89b609f3fee44899185bd5dc277ae5- Stable URL of this page
https://metraipolis.com/documento/dc_2a2a34e2c483
How to cite
Beyond Expected Values: Risk-Sensitive Planning with Distributional Monte-Carlo Tree Search — arXiv:2610.23565 — arXiv (en), https://arxiv.org/abs/2610.23565 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2a2a34e2c483 (Metraipolis archive dc_2a2a34e2c483, sha256 1683d54ac7d7314c4b347a5fd55a49ae6f89b609f3fee44899185bd5dc277ae5)
BibTeX
@misc{dc2a2a34e2c483,
title = {Beyond Expected Values: Risk-Sensitive Planning with Distributional Monte-Carlo Tree Search — arXiv:2610.23565},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.23565},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2a2a34e2c483; sha256 1683d54ac7d7314c4b347a5fd55a49ae6f89b609f3fee44899185bd5dc277ae5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23565. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Monte-Carlo Tree Search (MCTS) typically stores only scalar mean estimates, which makes it poorly suited for tail-risk objectives such as conditional value-at-risk (CVaR). We introduce Categorical Thompson Sampling with Optimism (CATSO) and Particle Thompson Sampling with Optimism (PATSO), two distributional MCTS algorithms that maintain finite-support empirical backup laws at Q-edges. CATSO represents Q-edge laws with categorical atoms and Dirichlet Thompson sampling, while PATSO represents them with adaptive particles. Both algorithms select actions using the CVaR of a Thompson-sampled Q-edge law plus a polynomial optimism bonus, and propagate scalar continuation values through visit-weighted averages of child Q-edge CVaR scores. For the finite-depth nested CVaR planning target, we prove an $O(n^{-1/2})$ bound on the root value error after $n$ rollouts, which upper-bounds the risk-sensitive simple regret. CATSO adds a fixed-grid discretization bias, while capped PATSO adds a tunable Wasserstein compression bias. Experiments on stochastic planning benchmarks show that distributional Q-edges improve lower-tail return and reduce catastrophic outcomes when mean-optimal and risk-sensitive behavior differ.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Machine Learning (stat.ML).
Pagina del lavoro: https://arxiv.org/abs/2610.23565 — 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 2026-10-19T09:06:29.555Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).