Unifying and Understanding Overparameterized Circuit Representations via Low-Rank Tensor Decompositions

Authors: Antonio Mari, Gennaro Vessio, Antonio Vergari

Project URL: https://openreview.net/forum?id=1btutFdIya

Paper: https://openreview.net/pdf?id=1btutFdIya

Abstract: Tensorizing probabilistic circuits (PCs) -- structured computational graphs capable of efficiently and accurately performing various probabilistic reasoning tasks -- is the go-to way to represent and learn these models. This paper systematically explores the architectural options employed in modern overparameterized PCs, namely RAT-SPNs, EiNets, and HCLTs, and unifies them into a single algorithmic framework. By trying to compress the existing overparameterized layers via low-rank decompositions, we discover alternative parameterizations that possess the same expressive power but are computationally more efficient. This emphasizes the possibility of “mixing & matching” different design choices to create new PCs and helps to disentangle the few ones that really matter.

@inproceedings{mari2023lorapc,
 title={Unifying and Understanding Overparameterized Circuit Representations via Low-Rank Tensor Decompositions},
 author={Antonio Mari, Gennaro Vessio, Antonio Vergari},
 booktitle={The 6th Workshop on Tractable Probabilistic Modeling},
 year={2023}}