Research map / Bandits and online learning / Online convex optimization and games
Modelling and Solving Online Optimisation Problems
Alexander Ek, Maria Garcia de la Banda, Andreas Schutt, Peter J. Stuckey, Guido Tack
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 |
|---|---|---|---|
| add garbage collection (ablation) | online MiniZinc job-shop · abz5, 300 iterations | accumulated instantiation time (s): 3,913.48 -> 19.00 | Improved |
| add garbage collection (ablation) | online MiniZinc job-shop · abz5, 300 iterations | accumulated solving time (s): 75.64 -> 13.01 | Improved |
| add garbage collection (ablation) | online MiniZinc job-shop · ft10, 300 iterations | accumulated instantiation time (s): 3,882.49 -> 19.86 | Improved |
| add garbage collection (ablation) | online MiniZinc job-shop · ft20, 300 iterations | accumulated solving time (s): 1,490.86 -> 1,467.56 | No clear effect |
| add garbage collection (ablation) | online MiniZinc job-shop · ft20, 300 iterations | accumulated instantiation time (s): 941.64 -> 18.84 | Improved |
| shorten per-iteration time limit from 1.00 s to 0.50 s | real-time online job-shop · ft20, 4 jobs/s | makespan: 117,973 -> 117,845 | Improved |
| shorten per-iteration time limit to 0.25 s | real-time online job-shop · ft20, 4 jobs/s | makespan | Hurt |
| increase per-iteration time limit from 1.00 s to 4.00 s | real-time online job-shop · ft20, 8 jobs/s | makespan: 70,663 -> 67,892 | Improved |
| increase per-iteration time limit from 4.00 s to 16.00 s | real-time online job-shop · ft20, 8 jobs/s | makespan: 67,892 -> 71,687 | Hurt |
| increase per-iteration time limit from 0.50 s to 16.00 s | real-time online job-shop · ft20, 4 jobs/s | makespan: 117,845 -> 121,352 | Hurt |
| use sliding-window decomposition instead of solving offline | cargo assembly planning · CAPP 07 | total delay: 10,563 -> 5,868 | Improved |
| use sliding-window decomposition instead of solving offline | cargo assembly planning · CAPP 19 | total delay: 141,119 -> 129,748 | Improved |
Closest papers by meaning
- Decoupling Learning and Decision-Making: Breaking the $\mathcal{O}(\sqrt{T})$ Barrier in Online Resource Allocation with First-Order Methods (ICML 2024 · similarity 0.82)
- PEARL: Solver-in-the-Loop Interactive Optimization Modeling from Natural Language (NeurIPS 2026 · similarity 0.82)
- Wait-Less Offline Tuning and Re-solving for Online Decision Making (ICML 2025 · similarity 0.81)
- Bayesian Design Principles for Offline-to-Online Reinforcement Learning (ICML 2024 · similarity 0.81)
- OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling (ICLR 2025 · similarity 0.81)
- The Stability of Online Algorithms in Performative Prediction (NeurIPS 2026 · similarity 0.81)
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.
