Abstract
AbstractSelecting the right tuning parameters for algorithms is a pravelent problem in machine learning that can significantly affect the performance of algorithms. Data-efficient optimization algorithms, such as Bayesian optimization, have been used to automate this process. During experiments on real-world systems such as robotic platforms these methods can evaluate unsafe parameters that lead to safety-critical system failures and can destroy the system. Recently, a safe Bayesian optimization algorithm, called SafeOpt, has been developed, which guarantees that the performance of the system never falls below a critical value; that is, safety is defined based on the performance function. However, coupling performance and safety is often not desirable in practice, since they are often opposing objectives. In this paper, we present a generalized algorithm that allows for multiple safety constraints separate from the objective. Given an initial set of safe parameters, the algorithm maximizes performance but only evaluates parameters that satisfy safety for all constraints with high probability. To this end, it carefully explores the parameter space by exploiting regularity assumptions in terms of a Gaussian process prior. Moreover, we show how context variables can be used to safely transfer knowledge to new situations and tasks. We provide a theoretical analysis and demonstrate that the proposed algorithm enables fast, automatic, and safe optimization of tuning parameters in experiments on a quadrotor vehicle.
Funder
Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung
Natural Sciences and Engineering Research Council of Canada
Connaught New Researcher Award
ETH Zurich
Publisher
Springer Science and Business Media LLC
Subject
Artificial Intelligence,Software
Reference51 articles.
1. Achiam, J., Held, D., Tamar, A., & Abbeel, P. (2017). Constrained policy optimization. In Proceedings of the international conference on machine learning (ICML), 1705.10528.
2. Akametalu, A. K., Kaynama, S., Fisac, J. F., Zeilinger, M. N., Gillula, J. H., & Tomlin, C. J. (2014). Reachability-based safe learning with Gaussian processes. In Proceedings of the IEEE conference on decision and control (CDC) (pp. 1424–1431).
3. Álvarez, M. A., Rosasco, L., & Lawrence, N. D. (2012). Kernels for vector-valued functions: A review. Foundations and Trends in Machine Learning, 4(3), 195–266.
4. Aswani, A., Gonzalez, H., Sastry, S. S., & Tomlin, C. (2013). Provably safe and robust learning-based model predictive control. Automatica, 49(5), 1216–1226.
5. Berkenkamp, F., Krause, A., Angela, P., & Schoellig (2016a). Bayesian optimization with safety constraints: Safe and automatic parameter tuning in robotics. arXiv:160204450 [csRO]
Cited by
51 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献