Research map / Learning theory and algorithms / Combinatorial and search optimization
Automatic Construction of Parallel Portfolios via Explicit Instance Grouping
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 |
|---|---|---|---|
| add dynamic instance transfer to the portfolio construction process (ablation) | PCIT versus PCRS · SAT-Single test set | PAR-10: 152 -> 119 | Helped |
| add dynamic instance transfer to the portfolio construction process (ablation) | PCIT versus PCRS · SAT-Multi test set | PAR-10: 1435 -> 1164 | Helped |
| add dynamic instance transfer to the portfolio construction process (ablation) | PCIT versus PCRS · TSP-Single test set | PAR-10: 31 -> 24 | Helped |
| add dynamic instance transfer to the portfolio construction process (ablation) | PCIT versus PCRS · TSP-Multi test set | PAR-10: 30 -> 24 | Helped |
| construct portfolios using explicit grouping with dynamic instance transfer rather than the strongest reported competing ACPP method | PCIT · SAT-Single test set | PAR-10: 146 -> 119 | Helped |
| construct portfolios using explicit grouping with dynamic instance transfer rather than the strongest reported competing ACPP method | PCIT · SAT-Multi test set | PAR-10: 1326 -> 1164 | Helped |
| construct portfolios using explicit grouping with dynamic instance transfer rather than the strongest reported competing ACPP method | PCIT · TSP-Single test set | PAR-10: 29 -> 24 | Helped |
| construct portfolios using explicit grouping with dynamic instance transfer rather than the strongest reported competing ACPP method | PCIT · TSP-Multi test set | PAR-10: 24 -> 24 | No clear effect |
| combine configurations from a rich design space into a parallel portfolio rather than use the sequential baseline | ACPP portfolios · SAT-Single test set | PAR-10: 238 -> 119 | Helped |
| combine configurations from a rich design space into a parallel portfolio rather than use the sequential baseline | ACPP portfolios · SAT-Multi test set | PAR-10: 2275 -> 1164 | Helped |
| combine configurations from a rich design space into a parallel portfolio rather than use the sequential baseline | ACPP portfolios · TSP-Single test set | PAR-10: 118 -> 24 | Helped |
| combine configurations from a rich design space into a parallel portfolio rather than use the sequential baseline | ACPP portfolios · TSP-Multi test set | PAR-10: 99 -> 24 | Helped |
Closest papers by meaning
- MILP-StuDio: MILP Instance Generation via Block Structure Decomposition (NeurIPS 2024 · similarity 0.81)
- Improved Approximation Algorithms for Clustered TSP and Subgroup Planning (AAAI 2025 · similarity 0.81)
- Parallel Beam Search Algorithms for Domain-Independent Dynamic Programming (AAAI 2024 · similarity 0.79)
- ACM-MILP: Adaptive Constraint Modification via Grouping and Selection for Hardness-Preserving MILP Instance Generation (ICML 2024 · similarity 0.79)
- UniAP: Unifying Inter- and Intra-Layer Automatic Parallelism by Mixed Integer Quadratic Programming (CVPR 2025 · similarity 0.78)
- Neural Solver Selection for Combinatorial Optimization (ICML 2025 · similarity 0.78)
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.