Probabilistic circuits (PCs) have gained prominence in recent years as a versatile framework for discussing probabilistic models that support tractable queries and are yet expressive enough to model complex probability distributions. Nevertheless, tractability comes at a cost: PCs are less expressive than neural networks. In this paper we introduce probabilistic neural circuits (PNCs), which strike a balance between PCs and neural nets in terms of tractability and expressive power. Theoretically, we show that PNCs can be interpreted as deep mixtures of Bayesian networks. Experimentally, we demonstrate that PNCs constitute powerful function approximators.

翻译：概率电路（PCs）近年来作为一种通用框架备受关注，它既能支持可解查询，又具有足够表现力来建模复杂概率分布。然而，可解性是有代价的：概率电路的表现力弱于神经网络。本文提出概率神经电路（PNCs），在可解性与表现力之间实现了概率电路与神经网络的平衡。理论上，我们证明概率神经电路可被解读为贝叶斯网络的深度混合模型。实验上，我们验证了概率神经电路是强大的函数逼近器。