Research map / Graph learning / Graph algorithms and clustering
A Study of Lagrangean Decompositions and Dual Ascent Solvers for Graph Matching
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use AMP with label factors and dual-ascent message passing | Graph matching solver · car, motor, worms | convergence / anytime performance | Helped |
| Use AMP rather than DD | Graph matching solver · graph flow | lower bound: -2849.96 vs -2840.29 | Hurt |
| Use AMP rather than DD | Graph matching solver · worms | runtime (s): 1471.95 vs 213.89 | Helped |
| Use DD rather than the proposed solvers | Graph matching solver · hotel | runtime (s): 0.42 | Helped |
| Use coupled original and inverse formulations (-C) | Graph matching solver · dense datasets | convergence robustness | Helped |
| Use inverse formulation (-I) on sparse graphs | Graph matching solver · sparse graphs | computational cost | Hurt |
| Use AMCF rather than HBP with a better message schedule | Graph matching solver · worms | time to similar duality gap: one-two orders of magnitude faster | Helped |
| Use the AMP schedule and label-factor relaxation rather than the network-flow-factor schedule | Equivalent graph-matching relaxations · graph matching benchmarks | convergence speed | Helped |
| Use DD's larger local subproblems | DD · graph flow | lower bound | Helped |
| Use the GM relaxation rather than stronger relaxations | Graph matching solver · car, motor, worms | lower-bound performance | Hurt |
Closest papers by meaning
- A Dual Ascent Framework for Lagrangean Decomposition of Combinatorial Problems (CVPR 2017 · similarity 0.91)
- A Convex Relaxation for Multi-Graph Matching (CVPR 2019 · similarity 0.90)
- Alternating Direction Graph Matching (CVPR 2017 · similarity 0.89)
- FRAM: Frobenius-Regularized Assignment Matching with Mixed-Precision Computing (NeurIPS 2025 · similarity 0.88)
- Learning Partial Graph Matching via Optimal Partial Transport (ICLR 2025 · similarity 0.86)
- Discrete Cycle-Consistency Based Unsupervised Deep Graph Matching (AAAI 2024 · similarity 0.85)
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.