AI Paper Atlas

Research map / Adversarial robustness and security / Certified and empirical robustness

Transferable Adversarial Robustness for Categorical Data via Universal Robust Embeddings

Klim Kireev, Maksym Andriushchenko, Carmela Troncoso, Nicolas Flammarion

NeurIPS 2023 · Adversarial robustness and security › Certified and empirical robustness · approach: Adversarial training with feature consistency · 1 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
apply universal robust embeddingsRandom Forest · IEEECISrobust accuracy (%): 48.0 -> 81.0Improved
apply universal robust embeddingsLGBM · IEEECISrobust accuracy (%): 53.9 -> 78.5Improved
apply universal robust embeddingsRandom Forest · BAFrobust accuracy (%): 42.8 -> 65.8Improved
apply universal robust embeddingsRandom Forest · Creditrobust accuracy (%): 55.2 -> 66.7Improved
apply universal robust embeddingsLGBM · IEEECISclean accuracy (%): 81.2 -> 79.3Hurt
apply universal robust embeddingsLGBM · Creditrobust accuracy (%): 69.9 -> 72.5Improved
use robust embeddings instead of l1-trained boosted stumps (GBS-W)Gradient Boosted Stumps · IEEECISrobust accuracy (%): 11.1 -> 66.3Improved
merge close embeddings (tau=0.15) after bilevel training (ablation)robust embeddings, GBS · BAFrobust accuracy (%): 48.3 -> 70.9Improved
merge close embeddings in random embeddings (ablation)random embeddings, GBS · BAFrobust accuracy (%): 49.1 -> 44.7No clear effect
bilevel optimization instead of normal adversarial training for embeddings (tau=0.15) (ablation)robust embeddings, GBS · BAFrobust accuracy (%): 53.3 -> 70.9Improved
use first-layer instead of last-layer embeddings (ablation)LGBM · IEEECISclean accuracy (%): 78.9 -> 81.0Improved
use FT-Transformer embeddings instead of TabNet embeddings (ablation)Gradient Boosted Stumps · BAFrobust accuracy (%): 67.7 -> 71.5Improved

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