Data poisoning attacks in contextual bandits
WebTrojAI Literature Review. The list below contains curated papers and arXiv articles that are related to Trojan attacks, backdoor attacks, and data poisoning on neural networks and machine learning systems. They are ordered "approximately" from most to least recent and articles denoted with a "*" mention the TrojAI program directly. WebWe study offline data poisoning attacks in contextual bandits, a class of reinforcement learning problems with important applications in online recommendation and adaptive medical treatment, among others. We provide a general attack framework …
Data poisoning attacks in contextual bandits
Did you know?
WebIn this paper, we study the action poisoning attack against linear contextual bandit in both white-box and black-box settings. In the white-box setting, we assume that the attacker knows the coefficient vectors associated with arms. Thus, at each round, the attacker knows the mean rewards of all arms. While it is often unrealistic to exactly know WebAug 27, 2024 · For example, you can use a contextual bandit to select which news article to show first on the main page of your website to optimize click through rate. The context is information about the user: where they come from, previously visited pages of the site, device information, geolocation, etc. An action is a choice of what news article to display.
WebData Poisoning Attacks in Contextual Bandits 3 Formally, a contextual bandit has a set Xof contexts and a set A= f1;2;:::;Kgof K arms. A contextual bandit algorithm proceeds … Webon when and where the attack happens. In a typical data poisoning attack (a.k.a. training-time attack) setting, the attacker tampers the training data during training time to downgrade the utility of the learned model. On the other hand, in adver-sarial examples (a.k.a test-time attack), the attacker manipulates features of a target
WebFeb 10, 2024 · In this paper, we study several attack scenarios and show that a malicious agent can force a linear contextual bandit algorithm to pull any desired arm T - o (T) … WebMay 16, 2024 · Stochastic multi-armed bandits form a class of online learning problems that have important applications in online recommendation systems, adaptive medical treatment, and many others. Even though potential attacks against these learning algorithms may hijack their behavior, causing catastrophic loss in real-world applications, little is known ...
WebData Poisoning, Backdoor Attacks, and Defenses Micah Goldblum*1, Dimitris Tsipras2, ... Contextual bandits, often used in adaptive medical treatment, can be manipulated by …
WebY. Ma, K.-S. Jun, L. Li, and J. Zhu: Data poisoning attacks in contextual bandits. In the 9th Conference on Decision and Game Theory for Security (GameSec), ... L. Li, W. Chu, J. Langford, and R.E. Schapire: A contextual-bandit approach to personalized news article recommendation. In the 19th International Conference on World Wide Web ... coffee shop ghent norfolkWebApr 20, 2024 · Data poisoning attacks in contextual bandits. In International Conference on Decision and Game Theory for Security. Springer, 186–204. Google Scholar Cross Ref; Yao Ma, Suhang Wang, Charu C Aggarwal, and Jiliang Tang. 2024. Graph convolutional networks with eigenpooling. In Proceedings of the 25th ACM SIGKDD International … coffee shop gatlinburg tnWebData poisoning attacks in contextual bandits. In Conference on Decision and Game Theory for Security (GameSec), 2024. Google Scholar Cross Ref; Ng, Andrew Y., Harada, Daishi, and Russell, Stuart J. Policy invariance under reward transformations: Theory and application to reward shaping. coffee shop georgetown seattleWebAug 17, 2024 · We study offline data poisoning attacks in contextual bandits, a class of reinforcement learning problems with important applications in online recommendation … coffee shop george streetWebFeb 10, 2024 · In this paper, we study several attack scenarios and show that a malicious agent can force a linear contextual bandit algorithm to pull any desired arm T - o (T) times over a horizon of T steps, while applying adversarial modifications to either rewards or contexts that only grow logarithmically as O (log T). We also investigate the case when a ... coffee shop gift basketsWebWe study offline data poisoning attacks in contextual bandits, a class of reinforcement learning problems with important applications in online recommendation and adaptive … camera the size of a grain of sandWebUpload an image to customize your repository’s social media preview. Images should be at least 640×320px (1280×640px for best display). camera thermique seek thermal