AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Association Graph Learning for Multi-Task Classification with Category Shifts

Jiayi Shen, Zehao Xiao, Xiantong Zhen, Cees G. M. Snoek, Marcel Worring

NeurIPS 2022 · Graph learning › Graph self-supervised learning · approach: Semi-supervised node representation learning · 2 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 assignment entropy maximization (L_AE) (ablation)association graph model · Office-Home, 75% missingharmonic mean H: 58.26 -> 60.59Improved
add assignment entropy maximization (L_AE) (ablation)association graph model · Office-Home, 75% missingA_m: 44.65 -> 47.51Improved
add assignment entropy maximization (L_AE) (ablation)association graph model · Office-Home, 75% missingA_o: 87.59 -> 87.16No clear effect
add assignment entropy maximization (L_AE) (ablation)association graph model · Office-Home, 75% missingaverage assignment entropy: 0.13 -> 0.84Improved
increase neighbor size from 0 to maximum 70 (ablation)association graph model · Office-Home, 75% missingharmonic mean H: 49.82 -> 60.59Improved
increase neighbor size from 8 to 70 (ablation)association graph model · Office-Home, 75% missingharmonic mean H: 58.45 -> 60.59Improved
increase neighbor size from 0 to 70 (ablation)association graph model · Office-Home, 75% missingA_m: 36.84 -> 47.51Improved
add message-passing layers (L=0 to L=4) (ablation)association graph model · Office-HomeaccuracyImproved
replace attention graph with association graph (type-specific metric functions) (ablation)graph model · Office-HomeaccuracyImproved
use association graph instead of ERM baselineOurs vs ERM · Office-Home, 75% missingharmonic mean H: 49.32 -> 60.59Improved
use association graph instead of second-best WeighLossesOurs vs WeighLosses · Office-Home, 75% missingharmonic mean H: 50.26 -> 60.59Improved
use association graph when no categories are missing (0%)Ours vs WeighLosses · Office-Home, 0% missingaccuracy A_o: 81.78 -> 82.01Improved

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