Metraipolis › archive › dc_20929a89dc45
Learning-Guided Planning in Large Dynamic Action Spaces: Budgeted Tree Search for One-to-Many Mobile Charging — arXiv:2609.17429
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.17429
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T18:38:57.775Z
- SHA-256 of the archived text
16319ee3faa5ab1f93429a3163abd9692c4f3ac20187294be394067e85671e32- Stable URL of this page
https://metraipolis.com/documento/dc_20929a89dc45
How to cite
Learning-Guided Planning in Large Dynamic Action Spaces: Budgeted Tree Search for One-to-Many Mobile Charging — arXiv:2609.17429 — arXiv (en), https://arxiv.org/abs/2609.17429 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_20929a89dc45 (Metraipolis archive dc_20929a89dc45, sha256 16319ee3faa5ab1f93429a3163abd9692c4f3ac20187294be394067e85671e32)
BibTeX
@misc{dc20929a89dc45,
title = {Learning-Guided Planning in Large Dynamic Action Spaces: Budgeted Tree Search for One-to-Many Mobile Charging — arXiv:2609.17429},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.17429},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_20929a89dc45; sha256 16319ee3faa5ab1f93429a3163abd9692c4f3ac20187294be394067e85671e32},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17429. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Many learned sequential decision systems map the current state directly to an action. That shortcut becomes brittle when candidate actions are numerous, geometrically structured, and rebuilt with the state. One-to-many mobile charging makes this setting concrete: with N=250 sensors, the initial state induces about 1,125 candidate charging-stop actions; each chosen stop simultaneously serves its in-range sensors, and the action universe changes as sensors die. LP-BTS is a learning-guided planning architecture: a graph proposal policy concentrates a small candidate support, a learned value critic evaluates leaves, and edge-budgeted PUCT compares short simulated futures before committing an action. Because the policy scores this set without a fixed output head, a single frozen checkpoint covers every evaluated setting, spanning action universes from 736 to 2,813 stops. Matched ablations reveal complementary effects: uniform sampling costs 8.8 survival percentage points, while, with targeted support fixed, PUCT jointly retains 1.4 points (about 3.5 of 250 sensors) and direct policy selection travels 23% farther. On a prospectively specified, sealed 30-scenario confirmatory bank evaluated once, LP-BTS attains the highest observed survival (0.4545) and alive-AUC (0.8031). Its estimated survival advantage over the strongest domain-engineered comparator is +0.0066 (95% CI [-0.0037, +0.0184]), an unresolved difference, while it exceeds a deadline heuristic and two source-derived direct-policy reconstructions on every paired scenario. Both learned rows are trained, source-derived reconstructions of variants reported by Gong et al. In this setting, the results provide controlled evidence about learning-guided planning in a large, dynamic action space. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-16 in cs.AI). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.17429 — 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 2026-10-15T18:03:09.755Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-16?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).