PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
June 6, 20240 citationsOpen Access

On the Hardness of Probabilistic Neurosymbolic Learning

View Full Paper
JMJaron MaeneKU LeuvenVDVincent DerkinderenVIB-KU Leuven Center for MicrobiologyLRLuc De RaedtÖrebro University

Key Points

Key points are not available for this paper at this time.

Abstract

The limitations of purely neural learning have sparked an interest in probabilistic neurosymbolic models, which combine neural networks with probabilistic logical reasoning. As these neurosymbolic models are trained with gradient descent, we study the complexity of differentiating probabilistic reasoning. We prove that although approximating these gradients is intractable in general, it becomes tractable during training. Furthermore, we introduce WeightME, an unbiased gradient estimator based on model sampling. Under mild assumptions, WeightME approximates the gradient with probabilistic guarantees using a logarithmic number of calls to a SAT solver. Lastly, we evaluate the necessity of these guarantees on the gradient. Our experiments indicate that the existing biased approximations indeed struggle to optimize even when exact solving is still feasible.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Maene et al. (2024) studied this question.

synapsesocial.com/papers/68e65e3eb6db6435875ecfdchttps://doi.org/10.48550/arxiv.2406.04472
Ask AI
Helpful
Bookmark
Share
View Full Paper