AI Paper Atlas

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

HRN: A Holistic Approach to One Class Learning

Wenpeng Hu, Mengyu Wang, Qi Qin, Jinwen Ma, Bing Liu

NeurIPS 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 24 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
add H-regularization to NLL loss (ablation)HRN · MNIST/fMNIST/CIFAR-10AUC: 55.42 -> 80.92 (MNIST)Improved
add 2-norm instance normalization (ablation)NLL+Hreg · CIFAR-10AUC: 55.26 -> 71.32Improved
use 2-norm instance normalization instead of instance normalization of [79] (ablation)HRN · MNISTAUC: 92.58 -> 97.59Improved
replace H-regularization with L2 regularization (ablation)HRN · MNISTAUC: 97.59 -> 91.37Hurt
replace NLL with square loss in full HRN (ablation)HRN · CIFAR-10AUC: 71.32 -> 71.08No clear effect
square loss instead of NLL without regularization (ablation)one-class loss · MNISTAUC: 55.42 -> 76.67Improved
add 20% noisy training dataHRN · MNISTAUC: 97.59 -> 92.70Hurt
add 20% noisy training dataHRN · CIFAR-10AUC: 71.32 -> 70.36No clear effect
ImageNet pre-trained feature extractorHRN · CIFAR-10AUC: 71.27 -> 96.6Improved
HRN vs TQM baselineHRN · ThyroidF1: 75.27 -> 95.87Improved
HRN vs iForestHRN · ArrhythmiaF1: 86.43 -> 84.46Hurt

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