Research map / Learning theory and algorithms / Combinatorial and search optimization
Neural Learning of One-of-Many Solutions for Combinatorial Problems in Structured Output Spaces
Yatin Nandwani, Deepanshu Jindal, Mausam ., Parag Singla
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 SelectR instead of MinLoss | N-Queens · N-Queens Overall | accuracy: 77.28 -> 79.72 | Improved |
| Use SelectR instead of MinLoss | Futoshiki · Futoshiki Overall | accuracy: 75.18 -> 76.4 | Improved |
| Use SelectR instead of MinLoss | Sudoku · Sudoku Overall | accuracy: 82.59 -> 85.21 | Improved |
| Use SelectR instead of I-ExplR | N-Queens · N-Queens Overall | accuracy: 77.7 -> 79.72 | Improved |
| Use SelectR instead of I-ExplR | Futoshiki · Futoshiki Overall | accuracy: 76.33 -> 76.4 | No clear effect |
| Use SelectR instead of I-ExplR | Sudoku · Sudoku Overall | accuracy: 84.46 -> 85.21 | Improved |
| Explicitly handle multiplicity with SelectR rather than Unique-only training | N-Queens · N-Queens MS | accuracy: 66.85 -> 79.68 | Improved |
| Explicitly handle multiplicity with SelectR rather than Unique-only training | Futoshiki · Futoshiki MS | accuracy: 19.13 -> 71.57 | Improved |
| Explicitly handle multiplicity with SelectR rather than Unique-only training | Sudoku · Sudoku MS | accuracy: 66.39 -> 81.73 | Improved |
| Use SelectR rather than Unique-only training | Sudoku · Sudoku OS | accuracy: 89.19 -> 88.69 | Hurt |
| Use SelectR rather than I-ExplR | Futoshiki · Futoshiki OS | accuracy: 78.15 -> 78.01 | Hurt |
| Use SelectR rather than I-ExplR | N-Queens · N-Queens MS | accuracy: 79.46 -> 79.68 | Improved |
Closest papers by meaning
- Teaching Transformers to Solve Combinatorial Problems through Efficient Trial & Error (NeurIPS 2025 · similarity 0.89)
- UDC: A Unified Neural Divide-and-Conquer Framework for Large-Scale Combinatorial Optimization Problems (NeurIPS 2024 · similarity 0.87)
- PolyNet: Learning Diverse Solution Strategies for Neural Combinatorial Optimization (ICLR 2025 · similarity 0.86)
- Geometric Algorithms for Neural Combinatorial Optimization with Constraints (NeurIPS 2025 · similarity 0.86)
- Causal language modeling can elicit search and reasoning capabilities on logic puzzles (NeurIPS 2024 · similarity 0.86)
- Neural Solver Selection for Combinatorial Optimization (ICML 2025 · 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.
