Soft Learning Probabilistic Circuits

Soroush Ghandi, Benjamin Quost, Cassio P. de Campos

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

Samenvatting

Probabilistic Circuits (PCs) are prominent tractable probabilistic models, allowing for a wide range of exact inferences. This paper focuses on the main algorithm for training PCs, LearnSPN, a gold standard due to its efficiency, performance, and ease of use, in particular for tabular data. We show that LearnSPN is a greedy likelihood maximizer under mild assumptions. While inferences in PCs may use the entire circuit structure for processing queries, LearnSPN applies a hard method for learning them, propagating at each sum node a data point through one and only one of the children/edges as in a hard clustering process. We propose a new learning procedure named SoftLearn, that induces a PC using a soft clustering process. We investigate the effect of this learning-inference compatibility in PCs. Our experiments show that SoftLearn outperforms LearnSPN in many situations, yielding better likelihoods and arguably better samples. We also analyze comparable tractable models to highlight the differences between soft/hard learning and model querying.
Originele taal-2Engels
Titel12th International Conference on Probabilistic Graphical Models, 11-13 September 2024, De Lindenberg, Nijmegen, the Netherlands
RedacteurenJohan Kwisthout, Silja Renooij
UitgeverijPMLR
Pagina's273-294
StatusGepubliceerd - 2024
Evenement12th International Conference on Probabilistic Graphical Models - Nijmegen, Nederland
Duur: 11 sep. 202413 sep. 2024

Publicatie series

NaamProceedings of Machine Learning Research
Volume246

Congres

Congres12th International Conference on Probabilistic Graphical Models
Land/RegioNederland
StadNijmegen
Periode11/09/2413/09/24

Vingerafdruk

Duik in de onderzoeksthema's van 'Soft Learning Probabilistic Circuits'. Samen vormen ze een unieke vingerafdruk.

Citeer dit