AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Neuron Coverage-Guided Domain Generalization

Chris Xing Tian, Haoliang Li, Xiaofei Xie, Yang Liu, Shiqi Wang

TPAMI 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Optimization-based domain generalization · 41 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 gradient similarity loss to coverage loss (ablation)NCDG · PACS SSDG (Sketch source)accuracy: 45.6 -> 53.1Improved
add coverage maximization loss to gradient similarity loss (ablation)NCDG · PACS SSDG (Sketch source)accuracy: 45.4 -> 53.1Improved
add gradient similarity loss to coverage loss (ablation)NCDG · PACS SSDG (Cartoon source)accuracy: 76.3 -> 76.4No clear effect
add gradient similarity loss to coverage loss (ablation)NCDG · PACS SSDG (Photo source)accuracy: 43.6 -> 49.0Improved
add coverage loss to AugmixNCDG · CIFAR-10-C digital (avg)accuracy: 76.6 -> 79.2Improved
add NCDG loss to adversarial augmentation GUDGUD · Digits SSDG (avg)accuracy: 54.6 -> 57.3Improved
add NCDG loss to M-ADAM-ADA · Digits SSDG (avg)accuracy: 59.5 -> 62.5Improved
add NCDG regularization over vanilla training on original and intensity-reversed dataVanilla** · Digits SSDG (avg)accuracy: 69.4 -> 73.4Improved
add coverage loss to DeepLabv3+DeepLabv3+ · GTA5 to CityscapesmIoU: 29.7 -> 34.7Improved
add coverage loss to DeepLabv3+ with ISWRobustNet · GTA5 to CityscapesmIoU: 36.6 -> 37.2Improved
apply coverage loss on all ResNet-18 layer outputs instead of first conv only (ablation)NCDG · PACS SSDG (Sketch source)accuracy: 46.7 -> 53.1Improved
replace coverage loss with L1 regularization (ablation)NCDG · PACS SSDG (Sketch source)accuracy: 53.1 -> 46.1Hurt

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