Year: 2019
Type: conference-paper
Abstract: Can machine learning models be easily fooled? Despite the recent surge of interest in learned adversarial attacks in other domains, in the context of recommendation systems this question has mainly been answered using hand-engineered fake user profiles. This paper attempts to reduce this gap. We provide a formulation for learning to attack a recommender as a repeated general-sum game between two players, i.e., an adversary and a recommender oblivious to the adversary's existence. We consider the challenging case of poisoning attacks, which focus on the training phase of the recommender model. We generate adversarial user profiles targeting subsets of users or items, or generally the top-K recommendation quality. Moreover, we ensure that the adversarial user profiles remain unnoticeable by preserving proximity of the real user rating/interaction distribution to the adversarial fake user distribution. To cope with the challenge of the adversary not having access to the gradient of the recommender's objective with respect to the fake user profiles, we provide a non-trivial algorithm building upon zero-order optimization techniques. We offer a wide range of experiments, instantiating the proposed method for the case of the classic popular approach of a low-rank recommender, and illustrating the extent of the recommender's vulnerability to a variety of adversarial intents. These results can serve as a motivating point for more research into recommender defense strategies against machine learned attacks.
Institution: University of Minnesota
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
Awards: BIGDATA: F: DKA: Collaborative Research: High-Dimensional Statistical Machine Learning for Spatio-Temporal Climate Data, G1648129465, TWC: Medium: Collaborative: HIMALAYAS: Hierarchical Machine Learning Stack for Fine-Grained Analysis of Malware Domain Groups, G2501928946, G2692851364, BIGDATA: F: DKA: Collaborative Research: High-Dimensional Statistical Machine Learning for Spatio-Temporal Climate Data, THE CENTRAL OBJECTIVE OF THE PROPOSED PROJECT IS TO DEVELOP ACCURATE, ROBUST, AND SCALABLE METHODS FOR DETECTING PRECURSORS TO AVIATION SAFETY INCIDE, III: Medium: Collaborative Research: Bayesian Modeling and Inference for Quantifying Terrestrial Ecosystem Functions, RI: Small: Finding Patterns in Complex Data with Probablistic Graphical Models, G5930584660, G6073761981, G7227591233, G7403353761, Collaborative Research: Understanding Climate Change: A Data Driven Approach, G8025912458, CAREER: Combinatorial Online Learning and its Applications, EAGER: Collaborative Research: Learning Relations between Extreme Weather Events and Planet-Wide Environmental Trends
OpenAlex ID: W2972646741