Revisiting Adversarially Learned Injection Attacks Against Recommender Systems

TL;DR

This paper critically examines adversarial injection attacks on recommender systems, revealing that precise optimization significantly enhances attack effectiveness and highlighting transferability limitations, which informs better defense strategies.

Contribution

It provides an exact optimization approach for attack generation and analyzes transferability, addressing previous assumptions of perfect knowledge and suboptimal attack methods.

Findings

Exact optimization increases attack impact

Transferability of attacks has limitations

Insights for developing defenses against injection attacks

Abstract

Recommender systems play an important role in modern information and e-commerce applications. While increasing research is dedicated to improving the relevance and diversity of the recommendations, the potential risks of state-of-the-art recommendation models are under-explored, that is, these models could be subject to attacks from malicious third parties, through injecting fake user interactions to achieve their purposes. This paper revisits the adversarially-learned injection attack problem, where the injected fake user `behaviors' are learned locally by the attackers with their own model -- one that is potentially different from the model under attack, but shares similar properties to allow attack transfer. We found that most existing works in literature suffer from two major limitations: (1) they do not solve the optimization problem precisely, making the attack less harmful than it could be, (2) they assume perfect knowledge for the attack, causing the lack of understanding for realistic attack capabilities. We demonstrate that the exact solution for generating fake users as an optimization problem could lead to a much larger impact. Our experiments on a real-world dataset reveal important properties of the attack, including attack transferability and its limitations. These findings can inspire useful defensive methods against this possible existing attack.