AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Mixed Sum-Product Networks: A Deep Architecture for Hybrid Domains

Alejandro Molina, Antonio Vergari, Nicola Di Mauro, Sriraam Natarajan, Floriana Esposito, Kristian Kersting

AAAI 2018 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Latent variable marginalization and estimators · 81 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
MSPN (RDC, iso) instead of HBN-MMHChybrid density estimator · UCI hybrid benchmarks (14)avg test log-likelihood: wins 11/14Improved
replace histogram leaves with isotonic regression leaves (ablation)MSPN (RDC) · UCI anneal-Uavg test log-likelihood: -60.314 -> -38.312Improved
replace histogram leaves with isotonic regression leaves (ablation)MSPN (RDC) · UCI australianavg test log-likelihood: -17.891 -> -31.021Hurt
replace Gower distance clustering with RDC pipeline (iso) (ablation)MSPN · UCI hybrid benchmarksavg test log-likelihood: -38.836 -> -38.312 (anneal-U)No clear effect
MSPN instead of Dirichlet on simplex datadensity estimator · NIPS+LDA 50avg test log-likelihood: 245.008 -> 338.477Improved
MSPN instead of Poisson SPNdensity estimator · NIPSavg test log-likelihood: -227.74 -> -144.41Improved
add symbolic semantic codes to MNIST autoencodingMSPN · MNISTmarginal log-likelihood and accuracyImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy