AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

DROCC: Deep Robust One-Class Classification

Sachin Goyal, Aditi Raghunathan, Moksh Jain, Harsha Vardhan Simhadri, Prateek Jain

ICML 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Learning from unlabeled and positive-unlabeled data · 7 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 DROCC's manifold-based adversarial negative generation instead of DeepSVDDOne-class anomaly detection · ArrhythmiaF1-Score: 0.54 ± 0.01 -> 0.69 ± 0.02Improved
Use DROCC's manifold-based adversarial negative generation instead of DeepSVDDOne-class anomaly detection · ThyroidF1-Score: 0.73 ± 0.00 -> 0.78 ± 0.03Improved
Use DROCC's manifold-based adversarial negative generation instead of DeepSVDDOne-class anomaly detection · AbaloneF1-Score: 0.62 ± 0.01 -> 0.68 ± 0.02Improved
Use DROCC instead of DeepSVDDOne-class anomaly detection · ImageNet-10 French HornAUC: 53.06 ± 6.52 -> 76.97 ± 1.67Improved
Add a Mahalanobis distance learned from feature influence to the DROCC-LF objectiveDROCC-LF · MNIST 0 vs. 1 with half-zero negativesRecall at 3% FPR: 98.16% -> 99.4%Improved
Use outlier exposure with DROCC instead of DeepSADLimited-negative classification · MNIST 0 vs. 1 with half-zero negativesRecall at 3% FPR: 90.91% -> 98.16%Improved
Sample synthetic negatives randomly rather than generating them by gradient ascent (ablation)DROCC · CIFAR-10 DeerAUC: 73.62 ± 2.00 -> 56.09 ± 1.19Hurt
Sample synthetic negatives randomly rather than generating them by gradient ascent (ablation)DROCC · CIFAR-10 CatAUC: 67.13 ± 1.51 -> 67.33 ± 0.72No clear effect
Sample synthetic negatives randomly rather than generating them by gradient ascent (ablation)DROCC · CIFAR-10 FrogAUC: 74.43 ± 0.92 -> 74.82 ± 1.77No clear effect
Sample synthetic negatives randomly rather than generating them by gradient ascent (ablation)DROCC · CIFAR-10 AirplaneAUC: 81.66 ± 0.22 -> 79.67 ± 2.09Hurt
Augment CIFAR-10 images with flips and small rotations (ablation)GEOM · CIFAR-10 DeerAUC: 87.8 ± 0.2 -> 65.7 ± 1.7Hurt
Augment CIFAR-10 images with flips and small rotations (ablation)DROCC · CIFAR-10 DeerAUC: 73.6 ± 2.0 -> 71.1 ± 2.9No clear effect

Later papers that compared with this method

3 later accepted papers report DROCC as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

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