The main goal of reading paper is not just understanding it. Try to understand the key concept, but we need to get new ideas and research directions from the paper.
Paper information
- title: Multi-Step Adversarial Perturbations on Recommender Systems Embeddings
- authors: Anelli, Vito Walter, Alejandro Bellogín, Yashar Deldjoo, Tommaso Di Noia, and Felice Antonio Merra.
- venue: arXiv preprint arXiv:2010.01329 (2020).
- pdf link: pdf link
- github: implementation and dataset
- abstract:
Recommender systems (RSs) have attained exceptional performance in learning users’ preferences and helping them in finding the most suitable products. Recent advances in adversarial machine learning (AML) in the computer vision domain have raised interests in the security of state-of-the-art model-based recommenders. Recently, worrying deterioration of recommendation accuracy has been acknowledged on several state-of-the-art model-based recommenders (e.g., BPR-MF) when machine-learned adversarial perturbations contaminate model parameters. However, while the single-step fast gradient sign method (FGSM) is the most explored perturbation strategy, multi-step (iterative) perturbation strategies, that demonstrated higher efficacy in the computer vision domain, have been highly under-researched in recommendation tasks.
In this work, inspired by the basic iterative method (BIM) and the projected gradient descent (PGD) strategies proposed in the CV domain, we adapt the multi-step strategies for the item recommendation task to study the possible weaknesses of embedding-based recommender models under minimal adversarial perturbations. Letting the magnitude of the perturbation be fixed, we illustrate the highest efficacy of the multi-step perturbation compared to the single-step one with extensive empirical evaluation on two widely adopted recommender datasets. Furthermore, we study the impact of structural dataset characteristics, i.e., sparsity, density, and size, on the performance degradation issued by presented perturbations to support RS designer in interpreting recommendation performance variation due to minimal variations of model parameters.
Summary: problems to address, key ideas, quick results
presentation link
Questions about the paper?
What do you like?
What you don't like?
How to improve?
Any new ideas?
Reproducing results (if any)
The main goal of reading paper is not just understanding it. Try to understand the key concept, but we need to get new ideas and research directions from the paper.
Paper information
Recommender systems (RSs) have attained exceptional performance in learning users’ preferences and helping them in finding the most suitable products. Recent advances in adversarial machine learning (AML) in the computer vision domain have raised interests in the security of state-of-the-art model-based recommenders. Recently, worrying deterioration of recommendation accuracy has been acknowledged on several state-of-the-art model-based recommenders (e.g., BPR-MF) when machine-learned adversarial perturbations contaminate model parameters. However, while the single-step fast gradient sign method (FGSM) is the most explored perturbation strategy, multi-step (iterative) perturbation strategies, that demonstrated higher efficacy in the computer vision domain, have been highly under-researched in recommendation tasks.
In this work, inspired by the basic iterative method (BIM) and the projected gradient descent (PGD) strategies proposed in the CV domain, we adapt the multi-step strategies for the item recommendation task to study the possible weaknesses of embedding-based recommender models under minimal adversarial perturbations. Letting the magnitude of the perturbation be fixed, we illustrate the highest efficacy of the multi-step perturbation compared to the single-step one with extensive empirical evaluation on two widely adopted recommender datasets. Furthermore, we study the impact of structural dataset characteristics, i.e., sparsity, density, and size, on the performance degradation issued by presented perturbations to support RS designer in interpreting recommendation performance variation due to minimal variations of model parameters.
Summary: problems to address, key ideas, quick results
presentation link
Questions about the paper?
What do you like?
What you don't like?
How to improve?
Any new ideas?
Reproducing results (if any)