← 研踪 PaperTrailarXiv:2611.25826 · v1
A Sparse Low-Rank Biclique Decomposition for Graphs
首次提交 2026-11-19 17:42
版本更新 2026-11-19 17:42
雷达观测 2026-11-20 19:58
摘要
Low-rank plus sparse decompositions, such as robust PCA, typically model the low-rank component as dense and the sparse component as corruption. This viewpoint is poorly suited to graph adjacency matrices, where sparsity is structural and dense low-rank factors destroy the combinatorial and computational properties of the graph. We propose signed biclique decomposition (SBD), a sparse low-rank decomposition for integer-weighted graphs that represents the structured component as a sum of sparse signed binary rank-one factors, together with a sparse signed residual. The resulting representation remains discrete, interpretable, sparse, and compatible with standard linear algebra. We formulate fixed-rank and greedy SBD objectives and derive a gain-based greedy algorithm with theoretical guarantees. Empirically, SBD recovers block structure in noisy graphs, while its factorized representation enables efficient sparse matrix-matrix multiplication (SpMM). On real graphs with strong shared-neighborhood structure, SBD achieves compression factors above and SpMM speedups up to over cuSPARSE, suggesting a practical alternative to dense low-rank decompositions for graphs. Code is available at https://github.com/IAntoineV/sbd.
作者备注:Accepted at NeurIPS 2026. 29 pages, including references and appendices
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-19 17:42 · 最新观测版本
A Sparse Low-Rank Biclique Decomposition for Graphs
Low-rank plus sparse decompositions, such as robust PCA, typically model the low-rank component as dense and the sparse component as corruption. This viewpoint is poorly suited to graph adjacency matrices, where sparsity is structural and dense low-rank factors destroy the combinatorial and computational properties of the graph. We propose signed biclique decomposition (SBD), a sparse low-rank decomposition for integer-weighted graphs that represents the structured component as a sum of sparse signed binary rank-one factors, together with a sparse signed residual. The resulting representation remains discrete, interpretable, sparse, and compatible with standard linear algebra. We formulate fixed-rank and greedy SBD objectives and derive a gain-based greedy algorithm with theoretical guarantees. Empirically, SBD recovers block structure in noisy graphs, while its factorized representation enables efficient sparse matrix-matrix multiplication (SpMM). On real graphs with strong shared-neighborhood structure, SBD achieves compression factors above and SpMM speedups up to over cuSPARSE, suggesting a practical alternative to dense low-rank decompositions for graphs. Code is available at https://github.com/IAntoineV/sbd.
查看此版本解读 →查看此版本原文 ↗