AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Certifying Robustness to Programmable Data Bias in Decision Trees

Anna P. Meyer, Aws Albarghouthi, Loris D'Antoni

NeurIPS 2021 · Learning theory and algorithms › Prediction and decision losses · approach: Optimal decision tree learning · 0 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
increase label-flipping bias from 0.05% to 1% of training setAntidote-P · COMPAScertified robustness (%): 81.9 -> 3.0Hurt
increase missing-data bias from 0.05% to 0.4%Antidote-P · Adult Incomecertified robustness (%): 96.0 -> 60.9Hurt
switch from missing-data to label-flipping bias model at 1.0% biasAntidote-P · Drug Consumptioncertified robustness (%): 85.1 -> 7.1Hurt
use label-flipping instead of missing data at 0.4% biasAntidote-P · Adult Incomecertified robustness (%): 60.9 -> 34.8Hurt
use targeted bias model instead of untargeted label-flipping at 0.4%Antidote-P · COMPAScertified robustness (%): 20.6 -> 50.5Improved
use targeted missing-data model instead of untargeted at 1.0%Antidote-P · Adult Incomecertified robustness (%): 31.6Improved
combine missing and fake bias (0.1% each) instead of 0.1% label-flippingAntidote-P · COMPAScertified robustness (%): 71.5 -> 50.5Hurt
compare dataset: Drug Consumption vs Adult Income vs COMPAS under label-flipping 0.2%Antidote-P · three datasetscertified robustness (%): 94.5 vs 70.2 vs 47.8Mixed
White women vs Black people subgroup under label-flipping 0.4%Antidote-P · COMPAScertified robustness (%): 50.4 vs 0Mixed
White women vs White men under label-flipping 0.4%Antidote-P · COMPAScertified robustness (%): 50.4 vs 38.8Mixed
targeted bias on (Black, positive) at high bias levels, White vs Black test samplesAntidote-P · COMPAScertified robustness (%): 32.9 vs 0Mixed
targeted predicate (White, negative) instead of (Black, positive)Antidote-P · COMPASdemographic robustness gapImproved

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