New DiffLift framework learns graph liftings for topological neural networks

Researchers have developed DiffLift, a novel framework for learning graph liftings to hypergraphs and cellular complexes. This end-to-end approach uses learned vertex-level latent representations to identify and parameterize distributions over candidate higher-order cells for inclusion. Experiments demonstrate that DiffLift outperforms existing lifting methods on graph and node classification benchmarks, achieving gains of up to 45% over static liftings. AI

IMPACT Enhances the expressive power of graph neural networks by enabling end-to-end learning of higher-order structures.

RANK_REASON The cluster contains a research paper detailing a new framework for topological neural networks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• DiffLift
• Gotit.pub
• Hugging Face
• IArxiv
• Jorge Luiz Franco
• ScienceCast

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →