AI Paper Atlas

Research map / Learning theory and algorithms / Fairness

Causal Disentangled Anchor Learning for Scalable Fair Multi-view Clustering

Suyuan Liu, Shengfei Wei, Wenjing Yang, Shengju Yu, Siwei Wang, Xueqiong Li, Wenpeng Lu, Xinwang Liu

ICML 2026 · Learning theory and algorithms › Fairness · approach: Fairness-constrained classification

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 all three losses (Lorth, Lindep, Lsens) to the base dual-anchor reconstruction (ablation)CDAL · JaffeACC: 51.57 -> 96.24Improved
add all three losses (ablation)CDAL · HarACC: 38.45 -> 75.26Improved
add all three losses (ablation)CDAL · YaleBal: 0.00 -> 12.50Improved
add orthogonality loss Lorth alone (ablation)CDAL · JaffeACC: 51.57 -> 60.75Improved
add independence loss Lindep alone (ablation)CDAL · JaffeACC: 51.57 -> 50.85Hurt
add Lindep alone (ablation)CDAL · YaleACC: 74.27 -> 79.06Improved
add Lorth alone (ablation)CDAL · YaleACC: 74.27 -> 60.82Hurt
increase the number of anchors m beyond a small value (ablation)CDAL · six datasetsACCHurt
increase the number of anchors m (ablation)CDAL · six datasetsBalNo clear effect
use CDAL's linear-time linearized HSIC anchor scheme instead of prior fair multi-view methodsCDAL vs IMVC · Harrunning time (s): 1395.08 -> 96.84Improved
use CDAL instead of single-view SFD on ZafarCDAL vs SFD · ZafarNMI: 2.00 -> 91.31Improved

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