Adversarially Robust Optimization with Gaussian Processes

2018/10/25 by Ilija Bogunovic, Bogunovic, Ilija, Jonathan Scarlett +5 · 13 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · #FOS: Computer and information sciences #Gaussian Processes and Bayesian Inference #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine Learning and Data Classification #Spectroscopy Techniques in Biomedical and Chemical Research

paper · pdf · doi:10.48550/arxiv.1810.10775

openalex publication_date 2018/10/25 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/30

Abstract

In this paper, we consider the problem of Gaussian process (GP) optimization with an added robustness requirement: The returned point may be perturbed by an adversary, and we require the function value to remain as high as possible even after this perturbation. This problem is motivated by settings in which the underlying functions during optimization and implementation stages are different, or when one is interested in finding an entire region of good inputs rather than only a single point. We show that standard GP optimization algorithms do not exhibit the desired robustness properties, and provide a novel confidence-bound based algorithm StableOpt for this purpose. We rigorously establish the required number of samples for StableOpt to find a near-optimal point, and we complement this guarantee with an algorithm-independent lower bound. We experimentally demonstrate several potential applications of interest using real-world data sets, and we show that StableOpt consistently succeeds in finding a stable maximizer where several baseline methods fail.

Cited by

• Robust Bayesian Optimisation with Unbounded Corruptions
• Value-at-Risk Optimization with Gaussian Processes
• Noisy-Input Entropy Search for Efficient Robust Bayesian Optimization
• No-Regret Learning in Unknown Games with Correlated Payoffs
• Zeroth-Order Algorithms for Nonconvex Minimax Problems with Improved Complexities
• Stochastic Linear Bandits Robust to Adversarial Attacks
• Federated Bayesian Optimization via Thompson Sampling
• On Lower Bounds for Standard and Robust Gaussian Process Bandit\n Optimization
• Bayesian Optimisation with Formal Guarantees
• Adaptive Parameter Optimization in Gaussian Processes: A Comprehensive Study of Uncertainty Quantification and Dimensional Scaling
• A Domain-Shrinking based Bayesian Optimization Algorithm with Order-Optimal Regret Performance
• Adversarial Robustness Guarantees for Classification with Gaussian Processes
• Min-Max Optimization without Gradients: Convergence and Applications to Adversarial ML

Related

• Robust Satisficing Gaussian Process Bandits Under Adversarial Attacks
• Stable Bayesian Optimisation via Direct Stability Quantification
• Adversarial Robustness Guarantees for Gaussian Processes
• On Lower Bounds for Standard and Robust Gaussian Process Bandit\n Optimization
• Adversarial Robustness Guarantees for Classification with Gaussian Processes