Parameterized Complexity of Chordal Conversion for Sparse Semidefinite Programs with Small Treewidth
From MaRDI portal
Publication:6441637
arXiv2306.15288MaRDI QIDQ6441637
Publication date: 27 June 2023
Abstract: If a sparse semidefinite program (SDP), specified over matrices and subject to linear constraints, has an aggregate sparsity graph with small treewidth, then chordal conversion will frequently allow an interior-point method to solve the SDP in just time per-iteration. This is a significant reduction over the minimum time per-iteration for a direct solution, but a definitive theoretical explanation was previously unknown. Contrary to popular belief, the speedup is not guaranteed by a small treewidth in , as a diagonal SDP would have treewidth zero but can still necessitate up to time per-iteration. Instead, we construct an extended aggregate sparsity graph by forcing each constraint matrix to be its own clique in . We prove that a small treewidth in does indeed guarantee that chordal conversion will solve the SDP in time per-iteration, to -accuracy in at most iterations. For classical SDPs like the MAX--CUT relaxation and the Lovasz Theta problem, the two sparsity graphs coincide , so our result provide a complete characterization for the complexity of chordal conversion, showing that a small treewidth is both necessary and sufficient for time per-iteration. Real-world SDPs like the AC optimal power flow relaxation have different graphs with similar small treewidths; while chordal conversion is already widely used on a heuristic basis, in this paper we provide the first rigorous guarantee that it solves such SDPs in time per-iteration. [Supporting code at https://github.com/ryz-codes/chordalConv/]
Has companion code repository: https://github.com/ryz-codes/chordalconv
This page was built for publication: Parameterized Complexity of Chordal Conversion for Sparse Semidefinite Programs with Small Treewidth
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6441637)