Revisiting Injective Attacks on Recommender Systems

Authors: Haoyang LI, Shimin DI, Lei Chen

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on three real-world datasets demonstrate the effectiveness of our proposed attacker.
Researcher Affiliation Academia 1Computer Science and Engineering, HKUST, Hong Kong, China 2Data Science and Analytics, HKUST(GZ), Guangzhou, China
Pseudocode Yes Thus, we propose a gradient-based greedy algorithm to maximize the objective in Alg. 1 in Appx. A.3.
Open Source Code No Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [No]
Open Datasets Yes We evaluate our attacker on widely used three real-world datasets: Movie Lens-100K (ML-100K) [14], Movie Lens-1M (ML-1M) [14], and Gowalla [5].
Dataset Splits No The paper discusses training of the target RecSys and the attacker, and mentions hyperparameter settings, but does not explicitly provide specific training/validation/test dataset splits with percentages or counts for reproducibility.
Hardware Specification No The paper does not explicitly mention specific hardware details such as GPU or CPU models used for the experiments within its main text or appendices.
Software Dependencies No The paper does not provide specific software dependencies with version numbers (e.g., library names like PyTorch with a version number).
Experiment Setup Yes For default, we set β1 = 0.01 for all datasets. For unnoticeable, we set the default to be similar to the average number of interacted items of real users as shown in Tab. 4, i.e., β2 is {0.05, 0.05, 0.002} for ML-100K, ML-1M, and Gowalla, respectively. The details about hyper-parameter settings are listed in Appx. B.2.