AI Paper Atlas

Research map / Graph learning / GNN architectures

On Dropping Clusters to Regularize Graph Convolutional Neural Networks

Xikun Zhang, Chang Xu, Dacheng Tao

ECCV 2020 · Graph learning › GNN architectures · approach: Deep graph convolution and simplification · 4 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 spatial correlation (neighbor-hop dropping) to depth-wise dropping (ablation)DropCluster · Northwestern-UCLATop-1 accuracy: 70.8% -> 72.2%Improved
add depth-wise correlation (correlated channels) to spatial dropping (ablation)DropCluster · Northwestern-UCLATop-1 accuracy: 71.1% -> 72.2%Improved
add depth-wise correlation to spatial dropping (ablation)DropCluster · Northwestern-UCLATop-5 accuracy: 99.1% -> 99.6%Improved
replace dropout with DropClusterST-GCN-U · Northwestern-UCLATop-1 accuracy: 70.5% -> 72.2%Improved
replace SpatialDropout with DropClusterST-GCN-U · Northwestern-UCLATop-1 accuracy: 70.8% -> 72.2%Improved
add DropCluster vs no droppingST-GCN · NTU-RGB+D X-Subaccuracy: 80.6% -> 83.3%Improved
add DropCluster vs no droppingST-GCN · NTU-RGB+D X-Viewaccuracy: 88.0% -> 88.9%Improved
replace Gaussian dropout with DropClusterST-GCN · NTU-RGB+D X-Subaccuracy: 82.2% -> 83.3%Improved
add JumpoutST-GCN · NTU-RGB+D X-Viewaccuracy: 88.0% -> 87.9%No clear effect
replace dropout with DropCluster at depth 11GCN · NTU-RGB+D X-View (11-layer net)accuracy: 88.6% -> 90.0%Improved
add DropCluster to GECNNGECNN · NTU-RGB+D X-Subaccuracy: 83.6% -> 84.7%Improved
add DropCluster to SLHMSLHM · NTU-RGB+D X-Subaccuracy: 84.3% -> 85.3%Improved

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