AI Paper Atlas

Research map / Optimization / Nonconvex and smooth optimization

Homeomorphic Projection to Ensure Neural-Network Solution Feasibility for Constrained Optimization

Enming Liang, Minghua Chen, Steven H. Low

JMLR 2024 · Optimization › Nonconvex and smooth optimization · approach: Complexity of nonconvex first-order methods · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add homeomorphic projection (H-Proj) as post-processingEq-NN-S and Eq-NN-U predictors · Convex QP/QCQP/SOCP/SDPfeasibility rate: 90.72 -> 100 (QP, Eq-NN-S); 49.32 -> 100 (QP, Eq-NN-U)Improved
add homeomorphic projection (H-Proj) as post-processingEq-NN-S predictor · AC-OPF 30/57/118-nodefeasibility rate: 77.73 -> 100 (30-node)Improved
replace gradient-based D-Proj with H-ProjEq-NN-S · QPfeasibility rate: 90.82 -> 100Improved
use H-Proj instead of standard orthogonal projection (Proj)Eq-NN-S · Convex QCQPaverage speedup: 21.3 -> 8353Improved
use H-Proj instead of standard orthogonal projection (Proj)Eq-NN-S · SOCPaverage speedup: 13.6 -> 6724Improved
use H-Proj instead of ProjEq-NN-S · SDPaverage speedup: 27.4 -> 369Improved
add H-Proj post-processingEq-NN-S · QPobjective error (all predictions): 0.45 -> 0.47Mixed
add H-Proj post-processingEq-NN-U · Convex QCQPobjective error (corrected infeasible only): 0.54 -> 0.78Hurt
add H-Proj post-processingEq-NN-S · 118-node AC-OPFobjective error (corrected infeasible only): 0.59 -> 2.44Hurt
use warm start (WS) instead of H-ProjEq-NN-S · QPspeedup and solution error: speedup 21.1 vs 184; solution error 1.46 vs 1.63Mixed
use 3-block INN rather than deeper (ablation)INN approximating MDH mapping · 2-D parametric constraint setconstraint approximationImproved
use infinity-norm ball instead of 2-norm ball as initial set (ablation)INN MDH mapping · 2-D parametric constraint setconstraint approximation qualityNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy