AI Paper Atlas

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

Are Labels Always Necessary for Classifier Accuracy Evaluation?

Weijian Deng, Liang Zheng

CVPR 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Group-robust classification · 79 citations (OpenAlex)

Check 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 neural network regression instead of linear regressionAutoEval · Digits (SVHN, USPS)RMSE (%): 9.87 -> 1.46Improved
Use neural network regression instead of linear regressionAutoEval · Natural images (PASCAL, Caltech, ImageNet)RMSE (%): 8.62 -> 3.04Improved
Increase the confidence threshold from τ=0.7 to τ=0.9Confidence score accuracy estimator · Natural images (PASCAL, Caltech, ImageNet)RMSE (%): 1.49 -> 6.96Hurt
Increase the confidence threshold from τ=0.7 to τ=0.9Confidence score accuracy estimator · Digits (SVHN, USPS)RMSE (%): 18.11 -> 25.59Hurt
Use confidence-threshold prediction instead of learned regressionAutoEval · Digits (SVHN, USPS)RMSE (%): 1.46 vs 18.11Hurt
Train the neural regressor with more meta-dataset sample sets (ablation)Neural network regression · Synthetic meta-datasetAbsolute prediction errorImproved
Increase the number of images in each meta-dataset sample set (ablation)Neural network regression · Synthetic meta-datasetAbsolute prediction errorImproved
Train the linear regressor with fewer meta-dataset sample sets (ablation)Linear regression · Synthetic meta-datasetAbsolute prediction errorNo clear effect
Apply unseen image transformations to real-world test datasets (ablation)Regression-based AutoEval · Edited PASCAL, Caltech, and ImageNetAbsolute prediction errorMixed

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