AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Variational Information Maximization for Feature Selection

Shuyang Gao, Greg Ver Steeg, Aram Galstyan

NeurIPS 2016 · Learning theory and algorithms › Prediction and decision losses · approach: Active learning and subset selection

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 variational mutual-information lower-bound forward selection with Naive Bayes Q-distributionfeature selection · 17 real-world datasetsaverage cross-validation error rateMixed
Use variational mutual-information lower-bound forward selection with pairwise Q-distributionfeature selection · 17 real-world datasetsaverage cross-validation error rateMixed
Use VMI_naive instead of mRMRfeature selection · Lungaverage cross-validation error rate: 10.9±(4.7) -> 7.4±(3.6)Improved
Use VMI_naive instead of SPEC_CMIfeature selection · Colonaverage cross-validation error rate: 16.1±(2.0) -> 11.2±(2.7)Improved
Use VMI_naive instead of mRMRfeature selection · Leukemiaaverage cross-validation error rate: 0.4±(0.7) -> 0.0±(0.1)Improved
Use VMI_naive instead of mRMRfeature selection · Lymphomaaverage cross-validation error rate: 5.6±(2.8) -> 3.7±(1.9)Improved
Use variational lower-bound forward selection rather than maximum relevance rankingsynthetic tree data · synthetic tree graphical modelfeature selection qualityMixed

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