# Safe non-smooth black-box optimization with application to policy search

arXiv: https://arxiv.org/abs/1912.09466 · HTML: https://mlsanity.com/paper/1912.09466/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Dec 2019.

## Built on by (most cited first, 6 of 6)

- [Safe Pontryagin Differentiable Programming](https://mlsanity.com/paper/2105.14937/) (2021)
- [Evading Black-box Classifiers Without Breaking Eggs](https://mlsanity.com/paper/2306.02895/) (2023)
- [Log Barriers for Safe Black-box Optimization with Application to Safe Reinforcement Learning](https://mlsanity.com/paper/2207.10415/) (2022)
- [A safe exploration approach to constrained Markov decision processes](https://mlsanity.com/paper/2312.00561/) (2023)
- [The Impact of the Geometric Properties of the Constraint Set in Safe Optimization with Bandit Feedback](https://mlsanity.com/paper/2305.00889/) (2023)
- [Safe Online Convex Optimization with Multi-Point Feedback](https://mlsanity.com/paper/2407.11471/) (2024)

## Abstract (arXiv)

For safety-critical black-box optimization tasks, observations of the constraints and the objective are often noisy and available only for the feasible points. We propose an approach based on log barriers to find a local solution of a non-convex non-smooth black-box optimization problem min f0(x) subject to fi(x)≤ 0,~ i = 1,ldots, m, at the same time, guaranteeing constraint satisfaction while learning an optimal solution with high probability. Our proposed algorithm exploits noisy observations to iteratively improve on an initial safe point until convergence. We derive the convergence rate and prove safety of our algorithm. We demonstrate its performance in an application to an iterative control design problem.