AI Paper Atlas

Research map / Optimization / Optimizers

Stochastic Optimization of Areas Under Precision-Recall Curves with Provable Convergence

Qi Qi, Youzhi Luo, Zhao Xu, Shuiwang Ji, Tianbao Yang

NeurIPS 2021 · Optimization › Optimizers · approach: Surrogate losses and robust solvers · 9 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 SOAP instead of cross-entropy trainingResNet18 · CIFAR-10 imbalancedtest AUPRC: 0.7155 -> 0.7629Improved
use SOAP instead of best baselineGINE/MPNN/ML-MPNN · HIVtest AUPRC: 0.3179 -> 0.3385 (GINE)Improved
use SOAP instead of best baselineResNet34 · Kaggle Melanomatest AUPRC: 0.2787 -> 0.3152Improved
use SOAP instead of best baselineGINE · MIT AICUREStest AUPRC: 0.5655 -> 0.6639Improved
AUROC maximization (AUC-M) instead of CEResNet34 · CIFAR-100 imbalancedtest AUPRC: 0.5792 -> 0.5306Hurt
swap squared hinge for logistic surrogate in SOAP (ablation)SOAP ResNet18 · CIFAR-10test AUPRC: 0.7629 -> 0.7542No clear effect
swap squared hinge for sigmoid surrogate in SOAP (ablation)SOAP ResNet18 · CIFAR-10test AUPRC: 0.7629 -> 0.7652No clear effect
swap squared hinge for logistic surrogate in SOAP (ablation)SOAP MPNN · HIVtest AUPRC: 0.3401 -> 0.3617Improved
lower class imbalance ratio (more extreme imbalance) (ablation)SOAP · Tox21/ToxCasttest AUPRCImproved
vary mini-batch size (ablation)SOAP · CIFAR-10/100test AUPRCNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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