Computational Optimization and Applications· 2026Q1
Reinforcement learning for adaptive interior point methods in convex quadratic programming
- 1citations
- Q1SCImago
- 2026year
Short summary
Reinforcement learning (RL) significantly speeds up regularized interior-point solvers for convex quadratic programming by automating hyperparameter tuning, achieving faster convergence to high-accuracy solutions after lightweight training.
AI-generated from the title and abstract; the full text is not read.
Key points
- Reinforcement learning (RL) is applied to optimize hyperparameter tuning in regularized interior-point solvers for convex quadratic programming.
- The RL approach addresses slow tail-convergence and improves solver performance, leading to high-accuracy solutions.
- Numerical experiments demonstrate that a lightweight RL training allows the learned policy to generalize to various problem classes and dimensions.
- The method facilitates solver tuning and speeds up the overall optimization process.
AI-generated from the title and abstract; the full text is not read.
Abstract
Abstract Quadratic programming is a workhorse of modern nonlinear optimization, control, and data science. Although regularized methods offer convergence guarantees under minimal assumptions on the problem data, they can exhibit the slow tail-convergence typical of first-order schemes, thus requiring many iterations to achieve high-accuracy solutions. Moreover, hyperparameter tuning significantly impacts the solver performance but how to find an appropriate parameter configuration remains an elusive research question. To address these issues, we explore how data-driven approaches can accelerate the solution process. Aiming at high-accuracy solutions, we focus on a regularized interior-point solver and carefully handle its two-loop flow and control parameters. We will show that reinforcement learning can make a significant contribution to facilitating the solver tuning and to speeding up the optimization process. Numerical experiments demonstrate that, after a lightweight training, the learned policy generalizes well to different problem classes with varying dimensions.
The authors' abstract, as published at the source. Computational Optimization and Applications, 2026 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Numerical Analysis
Numerical AnalysisMathematics