AI Paper Atlas

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

AUC Optimization with a Reject Option

Song-Qing Shen, Bin-Bin Yang, Wei Gao

AAAI 2020 · Learning theory and algorithms › Prediction and decision losses · approach: Data attribution and recourse · 6 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
lower reject cost d from 0.5 to 0.42 in AUCRO (ablation)AUCRO · diabetestest AUC: .8318 -> .9129Improved
lower reject cost d from 0.5 to 0.42 in AUCRO (ablation)AUCRO · germantest AUC: .7896 -> .8826Improved
lower reject cost d from 0.5 to 0.42 in AUCRO (ablation)AUCRO · splicetest AUC: .9243 -> .9810Improved
lower reject cost d from 0.5 to 0.42 in AUCRO (ablation)AUCRO · w8atest AUC: .9703 -> .9954Improved
lower reject cost d from 0.5 to 0.42 in AUCRO (ablation)AUCRO · covtypetest AUC: .8232 -> .9155Improved
AUCRO with no rejection (d=0.5) vs OAMseqAUCRO · connect-4test AUC: .7351 -> .8578Improved
AUCRO with no rejection (d=0.5) vs OAMgdAUCRO · covtypetest AUC: .7459 -> .8232Improved
AUCRO with no rejection (d=0.5) vs OPAUCAUCRO · diabetestest AUC: .8335 -> .8318No clear effect
AUCRO with no rejection (d=0.5) vs OPAUCAUCRO · germantest AUC: .7911 -> .7896No clear effect
AUCRO with no rejection (d=0.5) vs FSAUCAUCRO · w8atest AUC: .9482 -> .9703Improved
increase stepsize eta_t above 2^-4 (ablation)AUCRO (d=0.47) · Figure 2 datasetsAUCHurt
vary stepsize eta_t within [2^-12, 2^-4] (ablation)AUCRO (d=0.47) · Figure 2 datasetsAUCNo clear effect

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