Research map / Optimization / Nonconvex and smooth optimization
Lifted Inference for Convex Quadratic Programs
Martin Mladenov, Leonard Kleinhans, Kristian Kersting
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 |
|---|---|---|---|
| lift TC-QP-SVM via fractional symmetry (color refinement) | TC-QP-SVM · Cora | problem size: 78% of the variables, 70% of the constraints | Improved |
| lift TC-QP-SVM via fractional symmetry | TC-QP-SVM · Cora | accuracy | No clear effect |
| lift QP-SVM or LP-SVM (no transduction) | QP-SVM / LP-SVM · Cora | problem size | No clear effect |
| QP-SVM instead of LP-SVM | SVM on Cora · Cora | accuracy | Improved |
| add transductive citation constraints | QP-SVM / LP-SVM · Cora | accuracy | Improved |
| TC-QP-SVM instead of TC-LP-SVM | transductive SVM · Cora | accuracy | Improved |
| approximate lifting (LSVM) instead of ground SVM, centers 4 apart | RBF SVM · Gaussians | training time (secs): 20 -> 1.7 | Improved |
| approximate lifting (LSVM) instead of ground SVM, centers 4 apart | RBF SVM · Gaussians | test error: 0.02 -> 0.02 | No clear effect |
| approximate lifting (LSVM) instead of ground SVM, centers 2 apart | RBF SVM · Gaussians | training time (secs): 98 -> 2.1 | Improved |
| approximate lifting (LSVM) instead of ground SVM, centers 2 apart | RBF SVM · Gaussians | test error: 0.16 -> 0.17 | Hurt |
Closest papers by meaning
- Lifted Hinge-Loss Markov Random Fields (AAAI 2019 · similarity 0.85)
- Symmetry Guarantees Statistic Recovery in Variational Inference (NeurIPS 2026 · similarity 0.84)
- BayeSQP: Bayesian Optimization through Sequential Quadratic Programming (NeurIPS 2025 · similarity 0.83)
- Deep Distributed Optimization for Large-Scale Quadratic Programming (ICLR 2025 · similarity 0.83)
- Mirror Descent-Ascent for mean-field min-max problems (NeurIPS 2026 · similarity 0.83)
- Mirror Descent Beyond Euclidean Stability: An Exponential Separation in Initialization Sensitivity (NeurIPS 2026 · similarity 0.82)
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.
