AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Data Minimization at Inference Time

Cuong Tran, Ferdinando Fioretto

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Statistical analysis of ML practice · 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
Use personalized MinDRel feature disclosure instead of requiring all sensitive featuresInference on the evaluated learning tasks · Various benchmarksprediction accuracy: 10\% of their information while maintaining the same accuracy levelNo clear effect
Use F-Score selection rather than Importance or Random selection (ablation)MinDRel · Linear-classifier evaluationsdata leakageImproved
Increase the allowed failure probability δ (ablation)MinDRel · Linear-classifier evaluationsrelative data leakageImproved
Increase the allowed failure probability to δ ≤ 0.1 (ablation)MinDRel · Various datasets, nonlinear classifierssensitive features disclosed: 5\%Improved
Use MinDRel with δ = 0 rather than the all-features baselineNonlinear classifiers · Various datasetsaccuracy: 0.005\%Hurt
Use the Gaussian assumption rather than Bayesian NN to model conditional feature distributions (ablation)MinDRel · Bank, Income, Credit, Insurance (|S| = 5)training time (minutes): 204, 375, 125, 90 -> 0.01, 0.02, 0.02, 0.01Improved
Use the Gaussian assumption rather than Bayesian NN to model conditional feature distributions (ablation)MinDRel · Bank, Income, Credit, Insurance (|S| = 5, δ = 0)inference time on test set (minutes): 40, 254, 220, 34 -> 15, 78, 66, 9Improved
Use the Gaussian assumption rather than Bayesian NN to model conditional feature distributions (ablation)MinDRel · Credit dataset (|S| = 5)accuracy and data leakageNo clear effect
Reveal features in descending importance order rather than using F-Score (ablation)MinDRel · Linear-classifier evaluationsdata leakageHurt
Reveal features in random order rather than using F-Score (ablation)MinDRel · Linear-classifier evaluationsdata leakageHurt

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