AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Unsupervised Joint k-node Graph Representations with Compositional Energy-Based Models

Leonardo Cotta, Carlos H. C. Teixeira, Ananthram Swami, Bruno Ribeiro

NeurIPS 2020 · Graph learning › Graph self-supervised learning · approach: Semi-supervised node representation learning · 3 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
train MHM-GNN weights instead of random weights (ablation)MHM-GNN · Hyperedge Detection k=3, DBLPbalanced accuracy: 0.491 -> 0.773Improved
train MHM-GNN weights instead of random weights (ablation)MHM-GNN · Hyperedge Detection k=3, Corabalanced accuracy: 0.498 -> 0.618Improved
train MHM-GNN weights instead of random weights (ablation)MHM-GNN · Hyperedge Detection k=4, Pubmedbalanced accuracy: 0.510 -> 0.701Improved
train MHM-GNN weights instead of random weights (ablation)MHM-GNN · Hyperedge Detection k=5, Citeseerbalanced accuracy: 0.453 -> 0.703Improved
train MHM-GNN weights instead of random weights (ablation)MHM-GNN · DAG Leaf Counting k=3, Corabalanced accuracy: 0.497 -> 0.593Improved
train MHM-GNN weights instead of random weights (ablation)MHM-GNN · DAG Leaf Counting k=5, Pubmedbalanced accuracy: 0.244 -> 0.330Improved
train MHM-GNN weights instead of random weights (ablation)MHM-GNN · Hyperedge Detection k=5, Rent the Runwaybalanced accuracy: 0.924 -> 0.943No clear effect
use MHM-GNN joint k-node representation instead of GraphSAGE/DGI node representationsGraphSAGE, DGI · Hyperedge Detection k=3, Corabalanced accuracy: 0.490 -> 0.618Improved
use MHM-GNN instead of raw features on a task where structure is irrelevantRaw Features · Hyperedge Detection k=5, Rent the Runwaybalanced accuracy: 0.944 -> 0.943No clear effect
use MHM-GNN instead of GraphSAGE-meanGS-mean · DAG Leaf Counting k=4, Corabalanced accuracy: 0.245 -> 0.452Improved
use MHM-GNN instead of DGIDGI · Hyperedge Detection k=3, DBLPbalanced accuracy: 0.497 -> 0.773Improved
add node2vec/DeepWalk transductive embeddings with featuresnode2vec · Hyperedge Detection k=4, Corabalanced accuracy: 0.537 -> 0.626No clear effect

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