AI Paper Atlas

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

DisUnknown: Distilling Unknown Factors for Disentanglement Learning

Sitao Xiang, Yuming Gu, Pengda Xiang, Menglei Chai, Hao Li, Yajie Zhao, Mingming He

ICCV 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Group-robust classification · 6 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 unknown distillation (encoded unknown code plus feature matching loss) (ablation)DisUnknown · 3D Shapes, unknown = floor hueconsistency ratio: 63.42% -> 100.00%Improved
add unknown distillation (encoded unknown code plus feature matching loss) (ablation)DisUnknown · 3D Shapes, unknown = wall hueconsistency ratio: 55.63% -> 100.00%Improved
add unknown distillation (encoded unknown code plus feature matching loss) (ablation)DisUnknown · 3D Shapes, unknown = object hueconsistency ratio: 68.76% -> 100.00%Improved
use adversarial Stage II classifier instead of non-adversarial (ablation)DisUnknown · MNISTMIG: 0.916 -> 0.978Improved
use adversarial Stage II classifier instead of non-adversarial (ablation)DisUnknown · MNISTMSE: 0.0103 -> 0.0086Improved
add unknown code condition / code distance term (ablation)DisUnknown · MNISTMIG: 0.930 -> 0.978Improved
add unknown code condition / code distance term (ablation)DisUnknown · MNISTMSE: 0.0402 -> 0.0086Improved
add Negative Log Unlikelihood and sample-space classification in Stage I (ablation)DisUnknown · 3D Chairs, rotation unknowncode-space fluctuationImproved
merge more labeled factors into the unknown factor (ablation)DisUnknown · 3D Shapes, object hueMIG: 0.9501 -> 0.9868No clear effect
choose different factor as unknown (ablation)DisUnknown · 3D ShapesMIG: 0.9411 -> 0.9825No clear effect
add weak supervision (labeled factor) vs unsupervised methodsDisUnknown · MNISTMIG: 0.568 -> 0.978Improved

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