# Information Preserving Line Search via Bayesian Optimization

arXiv: https://arxiv.org/abs/2507.15485 · HTML: https://mlsanity.com/paper/2507.15485/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 41 references that are in the map.
- First posted on arXiv 21 Jul 2025.

## Builds on (most cited first, 9 of 9)

- [A Tutorial on Bayesian Optimization](https://mlsanity.com/paper/1807.02811/) (2018)
- [Bayesian Optimization with Gradients](https://mlsanity.com/paper/1703.04389/) (2017)
- [Practical and Rigorous Uncertainty Bounds for Gaussian Process Regression](https://mlsanity.com/paper/2105.02796/) (2021)
- [Probabilistic Line Searches for Stochastic Optimization](https://mlsanity.com/paper/1502.02846/) (2015)
- [GENO -- GENeric Optimization for Classical Machine Learning](https://mlsanity.com/paper/1905.13587/) (2019)
- [Optimization for Classical Machine Learning Problems on the GPU](https://mlsanity.com/paper/2203.16340/) (2022)
- [On the Equivalence of Automatic and Symbolic Differentiation](https://mlsanity.com/paper/1904.02990/) (2019)
- [LineWalker: Line Search for Black Box Derivative-Free Optimization and Surrogate Model Construction](https://mlsanity.com/paper/2307.10463/) (2023)
- [Efficient Line Search Method Based on Regression and Uncertainty Quantification](https://mlsanity.com/paper/2405.10897/) (2024)

## Abstract (arXiv)

Line search is a fundamental part of iterative optimization methods for unconstrained and bound-constrained optimization problems to determine suitable step lengths that provide sufficient improvement in each iteration. Traditional line search methods are based on iterative interval refinement, where valuable information about function value and gradient is discarded in each iteration. We propose a line search method via Bayesian optimization, preserving and utilizing otherwise discarded information to improve step-length choices. Our approach is guaranteed to converge and shows superior performance compared to state-of-the-art methods based on empirical tests on the challenging unconstrained and bound-constrained optimization problems from the CUTEst test set.