Visualizing and understanding Sum-Product Networks

Antonio Vergari, Nicola Di Mauro, Floriana Esposito

Research output: Contribution to journalArticlepeer-review

Abstract / Description of output

Sum-Product Networks (SPNs) are deep tractable probabilistic models by which several kinds of inference queries can be answered exactly and in a tractable time. They have been largely used as black box density estimators, assessed by comparing their likelihood scores on different tasks. In this paper we explore and exploit the inner representations learned by SPNs. By taking a closer look at the inner workings of SPNs, we aim to better understand what and how meaningful the representations they learn are, as in a classic Representation Learning framework. We firstly propose an interpretation of SPNs as Multi-Layer Perceptrons, we then devise several criteria to extract representations from SPNs and finally we empirically evaluate them in several (semi-)supervised tasks showing they are competitive against classical feature extractors like RBMs, DBNs and deep probabilistic autoencoders, like MADEs and VAEs.
Original languageEnglish
Pages (from-to)551-573
Number of pages23
JournalMachine Learning
Volume108
Issue number4
Early online date30 Aug 2018
DOIs
Publication statusPublished - 15 Apr 2019

Keywords / Materials (for Non-textual outputs)

  • Sum-product-networks
  • representation learning
  • tractable density estimation

Fingerprint

Dive into the research topics of 'Visualizing and understanding Sum-Product Networks'. Together they form a unique fingerprint.

Cite this