Research map / Learning theory and algorithms / Prediction and decision losses
Variational Information Maximization for Feature Selection
Shuyang Gao, Greg Ver Steeg, Aram Galstyan
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use variational mutual-information lower-bound forward selection with Naive Bayes Q-distribution | feature selection · 17 real-world datasets | average cross-validation error rate | Mixed |
| Use variational mutual-information lower-bound forward selection with pairwise Q-distribution | feature selection · 17 real-world datasets | average cross-validation error rate | Mixed |
| Use VMI_naive instead of mRMR | feature selection · Lung | average cross-validation error rate: 10.9±(4.7) -> 7.4±(3.6) | Improved |
| Use VMI_naive instead of SPEC_CMI | feature selection · Colon | average cross-validation error rate: 16.1±(2.0) -> 11.2±(2.7) | Improved |
| Use VMI_naive instead of mRMR | feature selection · Leukemia | average cross-validation error rate: 0.4±(0.7) -> 0.0±(0.1) | Improved |
| Use VMI_naive instead of mRMR | feature selection · Lymphoma | average cross-validation error rate: 5.6±(2.8) -> 3.7±(1.9) | Improved |
| Use variational lower-bound forward selection rather than maximum relevance ranking | synthetic tree data · synthetic tree graphical model | feature selection quality | Mixed |
Closest papers by meaning
- Revisiting Probability Distribution Assumptions for Information Theoretic Feature Selection (AAAI 2020 · similarity 0.92)
- A Rigorous Information-Theoretic Definition of Redundancy and Relevancy in Feature Selection Based on (Partial) Information Decomposition (JMLR 2023 · similarity 0.88)
- On Variational Bounds of Mutual Information (ICML 2019 · similarity 0.87)
- Flow-based Variational Mutual Information: Fast and Flexible Approximations (ICLR 2025 · similarity 0.87)
- Estimating Conditional Mutual Information for Dynamic Feature Selection (ICLR 2024 · similarity 0.85)
- Towards Generalization beyond Pointwise Learning: A Unified Information-theoretic Perspective (ICML 2024 · similarity 0.85)
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.
