EUVIMEDEUVIMED entdecken
Crossref

GammaZero: Learning to Guide Belief-Space Search for Long-Horizon POMDPs with Generalizable Graph Representations

Rajesh Mangannavar, Prasad Tadepalli

Proceedings of the International Conference on Automated Planning and Scheduling · 08.06.2026 · 36 (1) · 658-666

DOI: 10.1609/icaps.v36i1.42884 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the International Conference on Automated Planning and Scheduling, online erschienen am 08.06.2026. Verfasst von Rajesh Mangannavar und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We introduce an uncertainty-aware graph representation framework for learning to guide planning in Partially Observable Markov Decision Processes (POMDPs). Unlike existing approaches that require domain or problem size specific neural architectures, GammaZero leverages a unified graph-based belief representation that enables generalization across problem sizes within a domain. Our key insight is that belief states can be systematically transformed into uncertainty-aware graphs where structural patterns learned on small problems transfer to larger instances. We employ a graph neural network with a decoder architecture to learn value functions and policies from expert demonstrations on computationally tractable problems, then apply these learned heuristics to guide Monte Carlo tree search on larger problems. Experimental results on standard POMDP benchmarks demonstrate that GammaZero achieves comparable performance to BetaZero when trained and tested on the same-sized problems, while enabling zero-shot generalization to problems 2-6× larger than those seen during training.

Zitieren

APA

Mangannavar, R., & Tadepalli, P. (2026). GammaZero: Learning to Guide Belief-Space Search for Long-Horizon POMDPs with Generalizable Graph Representations. Proceedings of the International Conference on Automated Planning and Scheduling, 36(1), 658-666. https://doi.org/10.1609/icaps.v36i1.42884

Vancouver

Mangannavar R, Tadepalli P. GammaZero: Learning to Guide Belief-Space Search for Long-Horizon POMDPs with Generalizable Graph Representations. Proceedings of the International Conference on Automated Planning and Scheduling. 2026;36(1):658-666. doi:10.1609/icaps.v36i1.42884

BibTeX
@article{Mangannavar2026,
  author = {Mangannavar, Rajesh and Tadepalli, Prasad},
  title = {GammaZero: Learning to Guide Belief-Space Search for Long-Horizon POMDPs with Generalizable Graph Representations},
  journal = {Proceedings of the International Conference on Automated Planning and Scheduling},
  year = {2026},
  volume = {36},
  number = {1},
  pages = {658--666},
  doi = {10.1609/icaps.v36i1.42884}
}
Kostenlos starten →