Neural Bayesian Networks for Modular and Scalable Decision-Making Under Uncertainty

Abstract:Bayesian Networks (BNs) are established tools for letting agents reason under uncertainty in sequential decision-making. They split a joint distribution into local conditional mechanisms arranged on a directed acyclic graph, which makes them modular, interpretable, and ready to perform probabilistic inference. Their classical formulation, however, has limitations in both expressiveness and scale: each conditional is usually stored as a table or a linear-Gaussian form, and inference evaluates the graph one node and one query at a time. In this paper we introduce Neural Bayesian Networks (NBNs), in which each conditional is a differentiable, tensor-valued mechanism, while the graph and its factorization are unchanged. NBNs retain the modularity and local mechanisms of BNs, can represent any learnable conditional distribution, and run natively on GPUs. We also release NBN, a PyTorch-based library for learning and inference in NBNs, together with a public benchmark. Our evaluation shows gains in both expressiveness and scale: NBNs fit distributions that classical BN libraries struggle to represent, and on large networks and large batches of queries they run two orders of magnitude faster thanks to GPU parallelization. NBN is available at this https URL.

Main text 10 pages, extra is supplementary. ACM CCS: Computing methodologies~Artificial intelligence~Knowledge representation and reasoning~Probabilistic reasoning. Computing methodologies~Machine learning~Machine learning approaches~Neural networks. Computing methodologies~Machine learning~Machine learning approaches~Learning in probabilistic graphical models~Bayesian network models

Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

arXiv:2610.16902 [cs.AI]

(or arXiv:2610.16902v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2610.16902 arXiv-issued DOI via DataCite (pending registration)