Research map / Learning theory and algorithms / Combinatorial and search optimization
p-Box ADMM: A Versatile Framework for Integer Programming
Baoyuan Wu, Bernard Ghanem
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use l2-box ADMM instead of penalty method | binary quadratic program solver · UCI iris, wine, glass, letter clustering | objective: [0.69, 11.42, 1.16, 10.27]% | Improved |
| use l2-box ADMM instead of SDP relaxation | binary quadratic program solver · UCI iris, wine, glass clustering | objective: [1.41, 0, 1.19]% | Improved |
| run l2-box ADMM on GPU instead of CPU | l2-box ADMM · Cameraman segmentation (n=5x10^5) | runtime (s): 22.96 -> 7.12 | Improved |
| run on GPU instead of CPU at small size | l2-box ADMM · Cameraman segmentation (n=10^3) | runtime (s): 0.07 -> 0.1 | Hurt |
| use l2-box ADMM instead of min-cut | binary segmentation energy · Cameraman segmentation | energy: -3228 -> -3215 | Hurt |
| use l2-box ADMM instead of LP relaxation | energy minimization · Cameraman segmentation (n=5x10^5) | energy: -177932 -> -253121 | Improved |
| use l2-box ADMM instead of penalty method | energy minimization · Cameraman segmentation (n=10^3) | runtime (s): 84 -> 0.07 | Improved |
| use l2-box ADMM instead of IPFP_U and IFPU_S | graph matching · house video sequence | objective: 4% improvement | Improved |
| use l2-box ADMM instead of penalty method | graph matching · house video sequence | objective: 17% improvement | Improved |
| use l2-box ADMM instead of FGM | graph matching · house video sequence | objective | No clear effect |
| use l2-box ADMM instead of K-means | clustering · UCI wine | Rand Index (%): 93.57 -> 92.74 | Hurt |
| stop updating y2 after enough iterations (ablation) | l2-box ADMM · experiments | convergence speed | Improved |
Closest papers by meaning
- Adaptive Relaxed ADMM: Convergence Theory and Practical Implementation (CVPR 2017 · similarity 0.87)
- Optimizing ADMM and Over-Relaxed ADMM Parameters for Linear Quadratic Problems (AAAI 2024 · similarity 0.84)
- Accelerating Primal Solution Findings for Mixed Integer Programs Based on Solution Prediction (AAAI 2020 · similarity 0.84)
- Discerning the Linear Convergence of ADMM for Structured Convex Optimization through the Lens of Variational Analysis (JMLR 2020 · similarity 0.84)
- BIPNN: Learning to Solve Binary Integer Programming via Hypergraph Neural Networks (NeurIPS 2025 · similarity 0.84)
- SymILO: A Symmetry-Aware Learning Framework for Integer Linear Optimization (NeurIPS 2024 · similarity 0.84)
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.
