SyncAI.news, a Varaisys broadcasting
Graph Matching Relaxations and Amortization for Supervised Graph Prediction
FM

Federico M\'endez, Paul Krzakala, Gabriel Melo, Charlotte Laclau, R\'emi Flamary, Florence d'Alch\'e-Buc

· 1 min read

ResearcharXiv cs.LG

Graph Matching Relaxations and Amortization for Supervised Graph Prediction

arXiv:2609.15437v2 Announce Type: replace-cross Abstract: End-to-end Supervised Graph Prediction (SGP) requires a permutation-invariant loss to compare predicted and target graphs with arbitrary node orderings. Such losses typically involve a costly graph-matching problem. We first study three Optimal Transport relaxations of this problem and show, theoretically and empirically, that the Gromov-Wasserstein (GW) objective is the most suitable for SGP. Then, to avoid solving the resulting inner optimization for every training example, we propose to amortize the graph matching (node alignment) problem. For each training sample, the loss function leverages a transport plan provided by a parametric matcher based on the differentiable Sinkhorn algorithm applied on empirical node distributions. The graph prediction module and the matcher are jointly learned. We showcase the efficiency of this approach on toy and real world SGP problems of increasing complexity including a novel Mass-spectra to Scaffold task that we introduce.

Original source

This story was published by arXiv cs.LG and written by Federico M\'endez, Paul Krzakala, Gabriel Melo, Charlotte Laclau, R\'emi Flamary, Florence d'Alch\'e-Buc. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News