AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Zero-shot Transfer Learning within a Heterogeneous Graph via Knowledge Transfer Networks

Minji Yoon, John Palowitch, Dustin Zelle, Ziniu Hu, Russ Salakhutdinov, Bryan Perozzi

NeurIPS 2022 · Graph learning › Knowledge and temporal graphs · approach: Heterogeneous graph 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
add KTN domain adaptation to HMPNNHMPNN · OAG-CS P-A (L1)MRR: 0.297 -> 0.629Improved
add KTN to HMPNNHMPNN · OAG-CS V-A (L1)MRR: 0.133 -> 0.586Improved
add KTN to HMPNNHMPNN · PubMed G-DMRR: 0.354 -> 0.445Improved
add KTN to HANHAN · OAG-CS P-A (L1)MRR: 0.047 -> 0.497Improved
add KTN to MAGNNMAGNN · OAG-CS P-A (L1)MRR: 0.378 -> 0.556Improved
add KTN to R-GCNR-GCN · OAG-CS P-A (L1)NDCG: 0.337 -> 0.577Improved
add KTN to MPNNMPNN · OAG-CS A-P (L1)MRR: 0.145 -> 0.595Improved
replace KTN with DAN (MMD-based DA)HMPNN · OAG-CS P-A (L1)NDCG: 0.623 -> 0.452Hurt
replace MMD-based DA with adversarial DANNHMPNN · OAG-CS P-A (L1)NDCG: 0.452 -> 0.292Hurt
use label propagation (LP) instead of DAHMPNN · OAG-CS P-A (L1)MRR: 0.297 -> 0.041Hurt
shorten meta-path to length 1 (below minimum 2) (ablation)KTN · OAG-CS A-V (L1)MRR: 0.696 -> 0.010Hurt
lengthen meta-path from 2 to 3 (ablation)KTN · OAG-CS A-V (L1)MRR: 0.696 -> 0.648Hurt

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