Research map / Learning theory and algorithms / Combinatorial and search optimization
Learning Linear Programs from Optimal Decisions
Yingcong Tan, Daria Terekhov, Andrew Delong
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 gradient-based SQP outer solver (with LP differentiation) instead of gradient-free COBYLA | parametric LP learning · Synthetic PLP (D=10, M1=80, K=6) | fraction reaching zero AOE training loss: ~25\% -> 60-75\% | Improved |
| use gradient-based SQP instead of random search | parametric LP learning · Synthetic PLP (D=10, M1=80, K=6) | zero AOE training loss | Improved |
| gradient-based SQP instead of gradient-free methods for full coefficient learning (890 parameters) | parametric LP learning · Synthetic LP, all coefficients learned | zero AOE training loss | Improved |
| gradient-based SQP instead of RS and COBYLA | min-cost multi-commodity flow learning · Nguyen-Dupuis graph, 100 instances | training and testing loss | Improved |
| backprop through interior point instead of implicit or direct gradients (ablation) | SQP · Synthetic PLP | time to success | Hurt |
| implicit differentiation instead of backprop/direct (ablation) | SQP · Synthetic PLP with higher D | success rate | Improved |
Closest papers by meaning
- Inverse Optimization via Learning Feasible Regions (ICML 2025 · similarity 0.87)
- Solver-Free Decision-Focused Learning for Linear Optimization Problems (NeurIPS 2025 · similarity 0.87)
- Learning Generalized Linear Programming Value Functions (NeurIPS 2024 · similarity 0.87)
- What Data Enables Optimal Decisions? An Exact Characterization for Linear Optimization (NeurIPS 2025 · similarity 0.85)
- Decision-Focused Learning with Directional Gradients (NeurIPS 2024 · similarity 0.85)
- Multi-Stage Predict+Optimize for (Mixed Integer) Linear Programs (NeurIPS 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.
