AI Paper Atlas

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

Information criteria for non-normalized models

Takeru Matsuda, Masatoshi Uehara, Aapo Hyvarinen

JMLR 2021 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Likelihood-free and graphical model estimation · 7 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
use NCIC2 (parameter-count style correction) instead of NCIC1 (tr(IJ^-1) correction)NCIC bias correction · NCE with Gaussian mixture, N=10^3std of bias estimate: around 0.1 -> smaller than 10^{-8}Improved
use SMIC instead of NCIC1edge selection · GGM edge (1,2), N=100, sigma12=0.2detection probability: 0.515 -> 0.790Improved
use SMIC instead of AICedge selection · GGM edge (1,2), N=100, sigma12=0.2detection probability: 0.790 vs 0.783No clear effect
use NCIC1 instead of NCIC2edge selection · GGM edge (1,2), N=100, sigma12=0.2detection probability: 0.481 -> 0.515Improved
increase sigma12 from 0.2 to 0.5 (ablation)NCIC1 · GGM edge (1,2), N=100detection probability: 0.515 -> 0.943Improved
increase sample size from N=100 to N=200NCIC1 · GGM edge (1,2), sigma12=0.2detection probability: 0.515 -> 0.749Improved
use NCIC1 instead of NCE-CV leave-one-out cross-validationedge selection · Truncated GGM edge (1,2), N=100, sigma12=0.2detection probability: 0.210 -> 0.287Improved
use SMIC instead of SM-CV leave-one-out cross-validationedge selection · Truncated GGM edge (1,2), N=1000, sigma12=0.2detection probability: 0.599 -> 0.623No clear effect
use log-Gaussian graphical model instead of truncated Gaussian graphical modelSMIC model comparison · RNAseq data (d=40, N=487)SMICImproved
add dependence parameter lambda12 to bivariate von Mises model (ablation)NCIC2 model selection · Wind direction dataNCIC2: -1919 -> -1941Improved

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