AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Feature Grouping as a Stochastic Regularizer for High-Dimensional Structured Data

Sergul Aydore, Bertrand Thirion, Gael Varoquaux

ICML 2019 · Optimization › Implicit bias and generalization · approach: Neural tangent kernel analysis · 4 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 feature grouping instead of the strongest conventional MLP regularizerMLP on Olivetti Faces · No noisetest accuracy (%): 89.55 \pm 0.88 -> 85.45 \pm 1.08Hurt
Use feature grouping instead of the strongest conventional MLP regularizerMLP on HCP-small · No noisetest accuracy (%): 88.31 {\pm 0.14} -> 87.36 \pm 0.57Hurt
Use feature grouping instead of the strongest conventional MLP regularizerMLP on Olivetti Faces · Medium and high noisetest accuracy (%): 61.82 \pm 1.14 -> 80.91 {\pm 1.02}; 34.02 \pm 1.48 -> 68.79 {\pm 1.04}Improved
Use feature grouping instead of the strongest conventional MLP regularizerMLP on HCP-small · Medium and high noisetest accuracy (%): 80.66 \pm 0.22 -> 83.75 {\pm 0.35}; 69.81 \pm 0.56 -> 76.45 {\pm 0.52}Improved
Vary samples used per projection and number of projection matrices (ablation)Feature grouping with logistic regression · HCP-small, 330 samplesaccuracy: 79.88 \pm 0.53; 78.28 \pm 0.54; 79.24 \pm 0.66; 78.05 \pm 1.04; 78.82 \pm 0.79; 78.84 \pm 0.82; 77.97 \pm 0.823; 78.50 \No clear effect
Use feature grouping rather than conventional regularizers as the training sample count decreasesLogistic regression and MLP on HCP · Small-sample settingstest 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