Research map / Learning theory and algorithms / Combinatorial and search optimization
MIPaaL: Mixed Integer Program as a Layer
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 |
|---|---|---|---|
| Train with the exact MIP layer rather than the LP relaxation | MIPaaL · SP500 | monthly percentage increase: 1.97 $\pm$ 0.17 -> 2.79 $\pm$ 0.17 | Helped |
| Train with the exact MIP layer rather than separate prediction and optimization | MIPaaL · DAX | monthly percentage increase: 0.70 $\pm$ 1.46 -> 5.70 $\pm$ 0.68 | Helped |
| Train with the exact MIP layer rather than separate prediction and optimization | MIPaaL · Diverse Matching | successful matches: 3.42 $\pm$ 0.78 -> 4.80 $\pm$ 0.71 | Helped |
| Limit cut generation to 1000 cuts instead of generating cuts exactly (ablation) | MIPaaL · SP500 | monthly percentage increase: 2.79 $\pm$ 0.17 -> 2.60 $\pm$ 0.16 | Hurt |
| Limit cut generation to 1000 cuts instead of generating cuts exactly (ablation) | MIPaaL · DAX | monthly percentage increase: 5.70 $\pm$ 0.68 -> 4.39 $\pm$ 0.66 | Hurt |
| Limit cut generation to 1000 cuts instead of generating cuts exactly (ablation) | MIPaaL · Diverse Matching | successful matches: 4.80 $\pm$ 0.71 -> 3.45 $\pm$ 0.71 | Hurt |
| Limit cut generation to 100 cuts instead of generating cuts exactly (ablation) | MIPaaL · SP500 | monthly percentage increase: 2.79 $\pm$ 0.17 -> 1.25 $\pm$ 0.14 | Hurt |
| Limit cut generation to 100 cuts instead of generating cuts exactly (ablation) | MIPaaL · DAX | monthly percentage increase: 5.70 $\pm$ 0.68 -> 0.35 $\pm$ 0.63 | Hurt |
| Limit cut generation to 100 cuts instead of generating cuts exactly (ablation) | MIPaaL · Diverse Matching | successful matches: 4.80 $\pm$ 0.71 -> 2.57 $\pm$ 0.54 | Hurt |
Closest papers by meaning
- Melding the Data-Decisions Pipeline: Decision-Focused Learning for Combinatorial Optimization (AAAI 2019 · similarity 0.89)
- Solver-Free Decision-Focused Learning for Linear Optimization Problems (NeurIPS 2025 · similarity 0.88)
- Generalization Guarantees for Learning Score-Based Branch-and-Cut Policies in Integer Programming (NeurIPS 2025 · similarity 0.87)
- Contrastive Predict-and-Search for Mixed Integer Linear Programs (ICML 2024 · similarity 0.86)
- Fast and Interpretable Mixed-Integer Linear Program Solving by Learning Model Reduction (AAAI 2025 · similarity 0.86)
- 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.