Research map / Optimization / Nonconvex and smooth optimization
Quadratic Decomposable Submodular Function Minimization
Read the paperCheck 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 QRCD-SPE instead of the compared solvers | Hypergraph SSL solvers · Synthetic hypergraph | classification error | Helped |
| Use QRCD-SPE instead of the compared solvers | QDSFM solvers · Synthetic hypergraph | primal gap convergence | Helped |
| Use QRCD-SPE instead of PDHG and SGD | QDSFM solvers · Mushroom, Covertype45, and Covertype67 | convergence | Helped |
| Use QAP-SPE instead of QRCD-SPE | QDSFM solvers · Synthetic hypergraph | convergence within short CPU time | Hurt |
| Use PDHG or SGD instead of QRCD-SPE | QDSFM solvers · Synthetic hypergraph | convergence within short CPU time | Hurt |
| Use DRCD, which optimizes DSFM rather than QDSFM | Hypergraph SSL · Synthetic hypergraph | classification error | Hurt |
| Use InvLap instead of QRCD-SPE | Hypergraph SSL · Synthetic hypergraph | classification error | Hurt |
| Use InvLap on larger, fewer hyperedges | Hypergraph SSL · Mushroom, Covertype45, and Covertype67 | scalability | Hurt |
| Vary beta to small values (ablation) | QRCD · Synthetic hypergraph | convergence rate | Hurt |
| Increase beta above 1 (ablation) | QRCD · Synthetic hypergraph | convergence-rate dependence on beta | No clear effect |
| Vary the ratio max_{i,j} W_{ii}/W_{jj} (ablation) | QRCD · Synthetic hypergraph | convergence rate | Hurt |
| Replace exact projections with generic MNP or FW projections (ablation) | QRCD · Hypergraph SSL | runtime | Hurt |
Closest papers by meaning
- Decomposable Submodular Function Minimization via Maximum Flow (ICML 2021 · similarity 0.90)
- Decomposable Submodular Function Minimization: Discrete and Continuous (NeurIPS 2017 · similarity 0.90)
- Fast Decomposable Submodular Function Minimization using Constrained Total Variation (NeurIPS 2019 · similarity 0.89)
- Approximate Decomposable Submodular Function Minimization for Cardinality-Based Components (NeurIPS 2021 · similarity 0.89)
- From Linear to Linearizable Optimization: A Novel Framework with Applications to Stationary and Non-stationary DR-submodular Optimization (NeurIPS 2024 · similarity 0.87)
- Online Two-Stage Submodular Maximization (NeurIPS 2025 · similarity 0.87)
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.