AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Capacity and Bias of Learned Geometric Embeddings for Directed Graphs

Michael Boratko, Dongxu Zhang, Nicholas Monath, Luke Vilnis, Kenneth L. Clarkson, Andrew McCallum

NeurIPS 2021 · Graph learning › Graph self-supervised learning · approach: Heterophily and imbalance on graphs · 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
Use BOX embeddings instead of vector models at the lowest dimensionsBOX · Synthetic and real-world directed graphs, 8 params per nodeF1Improved
Use hyperbolic embeddings on tree-based graphsHyperbolic · nCRP and balanced treesF1Improved
Use hyperbolic embeddings on transitively closed graphsHyperbolic · Transitively closed graphsF1Hurt
Use vector-based models in low dimensionsSIM, COMPLEX, BILINEAR · Low-dimension graphsF1Hurt
Learn temperatures in box model (T-BOX) at higher dimensions (ablation)BOX · Synthetic and real-world graphs, 32 and 128 params per nodeF1Improved
Learn temperatures in box model with only 8 parameters per node (ablation)BOX · 8 params per nodeF1Hurt
Learn per-node and per-dimension temperature at fixed dimension (ablation)BOX · Varied dimensionalityF1Improved

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