Advances in Graph Learning and Optimization

The field of graph learning and optimization is rapidly advancing, with a focus on developing innovative methods for community detection, graph coarsening, and optimization on graph-structured data. Recent research has explored the integration of deep learning techniques with traditional rule-based constraints to improve community search, as well as the development of novel frameworks for graph contrastive learning and global optimization. Additionally, there has been a surge of interest in graph super-resolution, with new approaches being proposed to infer high-resolution graphs from low-resolution counterparts. Notable papers in this area include NCSAC, which introduces a novel approach for neural community search via attribute-augmented conductance, and HiLoMix, which proposes a robust high- and low-frequency graph learning framework for mixing address association. Other noteworthy papers include Dual-Kernel Graph Community Contrastive Learning, which transforms the input graph into a compact network of interconnected node sets, and Gromov-Wasserstein Graph Coarsening, which proposes two algorithms for graph coarsening within the Gromov-Wasserstein geometry.

Sources

NCSAC: Effective Neural Community Search via Attribute-augmented Conductance

Persistent reachability homology in machine learning applications

Optimal Parallel Basis Finding in Graphic and Related Matroids

Star-Based Separators for Intersection Graphs of $c$-Colored Pseudo-Segments

HiLoMix: Robust High- and Low-Frequency Graph Learning Framework for Mixing Address Association

Global Optimization on Graph-Structured Data via Gaussian Processes with Spectral Representations

Dual-Kernel Graph Community Contrastive Learning

Practical and Performant Enhancements for Maximization of Algebraic Connectivity

Gromov-Wasserstein Graph Coarsening

Rethinking Graph Super-resolution: Dual Frameworks for Topological Fidelity

Iterative Ricci-Foster Curvature Flow with GMM-Based Edge Pruning: A Novel Approach to Community Detection

Factorization-in-Loop: Proximal Fill-in Minimization for Sparse Matrix Reordering

A Spanning-Tree-Based Algorithm for Planar Graph Dismantling

Spatio-Temporal Graph Unlearning

Built with on top of