Adversarial Attacks on Gaussian Process Bandits
This work addresses security vulnerabilities in optimization algorithms for machine learning, but it is incremental as it builds on prior robustness studies.
This paper tackled the problem of adversarial attacks on Gaussian process bandits, specifically targeting the GP-UCB algorithm, by proposing various attack methods and demonstrating that these attacks can force the algorithm towards a target region with low attack budgets.
Gaussian processes (GP) are a widely-adopted tool used to sequentially optimize black-box functions, where evaluations are costly and potentially noisy. Recent works on GP bandits have proposed to move beyond random noise and devise algorithms robust to adversarial attacks. This paper studies this problem from the attacker's perspective, proposing various adversarial attack methods with differing assumptions on the attacker's strength and prior information. Our goal is to understand adversarial attacks on GP bandits from theoretical and practical perspectives. We focus primarily on targeted attacks on the popular GP-UCB algorithm and a related elimination-based algorithm, based on adversarially perturbing the function $f$ to produce another function $\tilde{f}$ whose optima are in some target region $\mathcal{R}_{\rm target}$. Based on our theoretical analysis, we devise both white-box attacks (known $f$) and black-box attacks (unknown $f$), with the former including a Subtraction attack and Clipping attack, and the latter including an Aggressive subtraction attack. We demonstrate that adversarial attacks on GP bandits can succeed in forcing the algorithm towards $\mathcal{R}_{\rm target}$ even with a low attack budget, and we test our attacks' effectiveness on a diverse range of objective functions.