EUVIMEDEUVIMED entdecken
Crossref

Paths Not Taken: Structure-Based Pruning in PSDD Learning and Inference

Cory Butz, Alejandro Santoscoy-Rivero, Camilla Lewis

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (24) · 19799-19807

DOI: 10.1609/aaai.v40i24.39062 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Cory Butz und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We make three novel contributions to parameter learning and inference in probabilistic sentential decision diagrams (PSDDs). First, rather than traversing the entire PSDD during parameter learning for each dataset example, we pioneer the use of determinism to focus only on the activated partition. Second, we demonstrate how to prune deterministic computation in inference, thereby eliminating the need to propagate probability over every node in the network for each query. Third, we introduce a technique that parallelizes a single circuit evaluation, rather than parallelizing individual multiplications or layer-wise inference. For both learning and inference, experimental results on benchmark PSDDs from various application domains demonstrate state-of-the-art performance.

Zitieren

APA

Butz, C., Santoscoy-Rivero, A., & Lewis, C. (2026). Paths Not Taken: Structure-Based Pruning in PSDD Learning and Inference. Proceedings of the AAAI Conference on Artificial Intelligence, 40(24), 19799-19807. https://doi.org/10.1609/aaai.v40i24.39062

Vancouver

Butz C, Santoscoy-Rivero A, Lewis C. Paths Not Taken: Structure-Based Pruning in PSDD Learning and Inference. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(24):19799-19807. doi:10.1609/aaai.v40i24.39062

BibTeX
@article{Butz2026,
  author = {Butz, Cory and Santoscoy-Rivero, Alejandro and Lewis, Camilla},
  title = {Paths Not Taken: Structure-Based Pruning in PSDD Learning and Inference},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {24},
  pages = {19799--19807},
  doi = {10.1609/aaai.v40i24.39062}
}
Kostenlos starten →