AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

CAT-Walk: Inductive Hypergraph Learning via Set Walks

Ali Behrouz, Farnoosh Hashemi, Sadaf Sadeghian, Margo Seltzer

NeurIPS 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Probabilistic embeddings and relational inference · 4 citations (OpenAlex)

Read the paper

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
Use CAT-Walk rather than the strongest listed baselineHyperedge prediction · Strongly inductive: all eight Table 1 datasetsAUC (%)Improved
Use CAT-Walk rather than the strongest listed baselineHyperedge prediction · Transductive: seven datasets other than Email EnronAUC (%)Improved
Use CAT-Walk rather than the strongest listed baselineHyperedge prediction · Email Enron, transductiveAUC (%): 83.59 ± 0.98 vs 80.45 ± 5.30Hurt
Add SetWalk instead of using a random walk (ablation)CAT-Walk · High School, inductiveAUC: 92.10 ± 2.18 -> 96.03 ± 1.50Improved
Add SetWalk instead of using a random walk (ablation)CAT-Walk · Primary School, inductiveAUC: 51.56 ± 5.63 -> 95.32 ± 0.89Improved
Add time encoding (ablation)CAT-Walk · Users in Threads, inductiveAUC: 70.58 ± 9.32 -> 89.84 ± 6.02Improved
Use SetMixer instead of Mean pooling (ablation)CAT-Walk · High School, inductiveAUC: 94.58 ± 1.22 -> 96.03 ± 1.50Improved
Use SetMixer instead of a sum-based universal set approximator (ablation)CAT-Walk · Question Tags U, inductiveAUC: 89.76 ± 0.45 -> 97.59 ± 2.21Improved
Use MLP-Mixer instead of RNN for walk encoding (ablation)CAT-Walk · Primary School, inductiveAUC: 50.29 ± 4.07 -> 95.32 ± 0.89Improved
Use nonzero temporal bias rather than fixing alpha = 0 (ablation)CAT-Walk · High School, inductiveAUC: 74.06 ± 14.9 -> 96.03 ± 1.50Improved
Add time encoding (ablation)CAT-Walk · Congress Bill, inductiveAUC: 92.56 ± 0.49 -> 93.54 ± 0.56Improved

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