AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Micro-Supervised Disturbance Learning: A Perspective of Representation Probability Distribution

Jielei Chu, Jing Liu, Hongjun Wang, Hua Meng, Zhiguo Gong, Tianrui Li

TPAMI 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Representation evaluation and trade-offs · 22 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 micro-supervised disturbance (SPI) to the representation model (Micro-DL vs NMicro-DL) (ablation)Micro-DL · 12 clustering datasets (average)clustering accuracy: 0.5619 -> 0.7343Improved
add micro-supervised disturbance (SPI) to the representation model (Micro-DL vs NMicro-DL) (ablation)Micro-DL · 12 clustering datasets (average)Jac index: 0.4015 -> 0.5836Improved
add micro-supervised disturbance (SPI) to the representation model (Micro-DL vs NMicro-DL) (ablation)Micro-DL · 12 clustering datasets (average)FM index: 0.5665 -> 0.7564Improved
add micro-supervised disturbance (SPI) to the representation model (Micro-DL vs NMicro-DL) (ablation)Micro-DL · 12 clustering datasets (average)Rand index: 0.5408 -> 0.5988Improved
use Micro-DL instead of Semi-SPclustering · 12 clustering datasets (average)clustering accuracy: increased by 0.3619Improved
use Micro-DL instead of Semi-MGclustering · 12 clustering datasets (average)clustering accuracy: 0.6692 -> 0.7343Improved
Micro-DL vs NMicro-DL significance (ablation)Micro-DL · Friedman + Nemenyi testp-value: 0.0023Improved
raise scale coefficient alpha from 0.1 to 0.3 (ablation)Micro-DL · 12 clustering datasetsclustering performanceImproved
raise scale coefficient alpha from 0.4 to 0.9 (ablation)Micro-DL · 12 clustering datasetsclustering performanceHurt

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