AI Paper Atlas

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

Verifying Robustness of Gradient Boosted Models

Gil Einziger, Maayan Goldstein, Yaniv Sa’ar, Itai Segall

AAAI 2019 · Adversarial robustness and security › Certified and empirical robustness · approach: Theory of adversarial risk · 49 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 smaller tree depth while keeping model accuracy similarGradient boosted models · MNIST, ε=1verified robust fraction (ρ): 53% -> 27.5%Improved
Increase tree depth from 4 to 20 in similarly accurate modelsGradient boosted models · MNIST, ε=1verified robust fraction (ρ): 53% -> 27.5%Hurt
Increase tree count from 50 to 100 at depth 8Gradient boosted model · MNIST, ε=1verified robust fraction (ρ): 53.5% -> 29%Mixed
Increase tree depth from 3 to 8 at 50 treesGradient boosted model · MNIST, ε=1verified robust fraction (ρ): 24% -> 53.5%Improved
Increase tree count from 20 to 100 at depth 3Gradient boosted model · MNIST, ε=1verified robust fraction (ρ): 16.5% -> 39.5%Improved
Increase tree count from 20 to 100 at depth 10Gradient boosted model · MNIST, ε=1verified robust fraction (ρ): 39.5% -> 15%Hurt
Increase tree depthGradient boosted regressors on HSKC · HSKCrobust fraction (ρ)Hurt
Increase number of treesGradient boosted regressors on HSKC · HSKCrobust fraction (ρ)Hurt
Increase perturbation bound εGradient boosted classifiers · MNISTverified robust fraction (ρ): 39.5% -> 10%Hurt

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