Fight Fire with Fire: Towards Robust Recommender Systems via Adversarial Poisoning Training

Wu, Chenwang; Lian, Defu; Ge, Yong; Zhu, Zhihao; Chen, Enhong; Yuan, Senchao

doi:10.1145/3404835.3462914

Cited by 26 publications

(16 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [16], the latent variable model is employed to reveal the user type by solving the program of maximization the entropy of ratings distribution in a star-rating system. In [26], the latent variable model relies on counteracting the adversarial poisoning attacks, which is the process of fake profiles generation that is intended to minimize the empirical risk. The algorithm described in [1] also uses the generative factor model, which utilizes the user metadata as an additional source of evidence against fake ratings.…”

Section: Related Workmentioning

confidence: 99%

Mitigating Human and Computer Opinion Fraud via Contrastive Learning

Tukmacheva¹,

Oseledets²,

Frolov³

2023

Preprint

View full text Add to dashboard Cite

We introduce the novel approach towards fake text reviews detection in collaborative filtering recommender systems. The existing algorithms concentrate on detecting the fake reviews, generated by language models and ignore the texts, written by dishonest users, mostly for monetary gains. We propose the contrastive learning-based architecture, which utilizes the user demographic characteristics, along with the text reviews, as the additional evidence against fakes. This way, we are able to account for two different types of fake reviews spamming and make the recommendation system more robust to biased reviews.

show abstract

Section: Related Workmentioning

confidence: 99%

Mitigating Human and Computer Opinion Fraud via Contrastive Learning

Tukmacheva¹,

Oseledets²,

Frolov³

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Multiple different RS have been proposed over the last years, many of them based on matrix factorization (MF) [35,25] or collaborative filtering (see, e.g., [37]). While recent methods increasingly rely on neural network-based factorization or recommendation (see, e.g., [42]), it remains debatable whether they yield superior results, e.g., with respect to performance and efficiency [21,30]. Factorization Machines (FM) represent another line of research which is closely related to MF.…”

Section: Related Literaturementioning

confidence: 99%

Factorized Structured Regression for Large-Scale Varying Coefficient Models

Rügamer¹,

Bender²,

Wiegrebe³

et al. 2022

Preprint

View full text Add to dashboard Cite

Recommender Systems (RS) pervade many aspects of our everyday digital life. Proposed to work at scale, state-of-the-art RS allow the modeling of thousands of interactions and facilitate highly individualized recommendations. Conceptually, many RS can be viewed as instances of statistical regression models that incorporate complex feature effects and potentially non-Gaussian outcomes. Such structured regression models, including time-aware varying coefficients models, are, however, limited in their applicability to categorical effects and inclusion of a large number of interactions. Here, we propose Factorized Structured Regression (FaStR) for scalable varying coefficient models. FaStR overcomes limitations of general regression models for large-scale data by combining structured additive regression and factorization approaches in a neural network-based model implementation. This fusion provides a scalable framework for the estimation of statistical models in previously infeasible data settings. Empirical results confirm that the estimation of varying coefficients of our approach is on par with state-of-the-art regression techniques, while scaling notably better and also being competitive with other time-aware RS in terms of prediction performance. We illustrate FaStR's performance and interpretability on a large-scale behavioral study with smartphone user data.

show abstract

“…Secondly, the AUC reduction on Electronics is smaller than that of Frappe. The possible reason is that Electronics has more data, corresponding to the finding that model robustness requires more data [50]. Finally, compared with FM, MixFM has a smaller reduction on AUC, revealing that using additional mixed data helps to improve FM's robustness.…”

Section: Robustness Evaluationmentioning

confidence: 99%

Boosting Factorization Machines via Saliency-Guided Mixup

Wu¹,

Lian²,

Ge³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Factorization machines (FMs) are widely used in recommender systems due to their adaptability and ability to learn from sparse data. However, for the ubiquitous non-interactive features in sparse data, existing FMs can only estimate the parameters corresponding to these features via the inner product of their embeddings. Undeniably, they cannot learn the direct interactions of these features, which limits the model's expressive power. To this end, we first present MixFM, inspired by Mixup, to generate auxiliary training data to boost FMs. Unlike existing augmentation strategies that require labor costs and expertise to collect additional information such as position and fields, these extra data generated by MixFM only by the convex combination of the raw ones without any professional knowledge support. More importantly, if the parent samples to be mixed have non-interactive features, MixFM will establish their direct interactions. Second, considering that MixFM may generate redundant or even detrimental instances, we further put forward a novel Factorization Machine powered by Saliency-guided Mixup (denoted as SMFM). Guided by the customized saliency, SMFM can generate more informative neighbor data. Through theoretical analysis, we prove that the proposed methods minimize the upper bound of the generalization error, which hold a beneficial effect on enhancing FMs. Significantly, we give the first generalization bound of FM, implying the generalization requires more data and a smaller embedding size under the sufficient representation capability. Finally, extensive experiments on five datasets confirm that our approaches are superior to baselines. Besides, the results show that "poisoning" mixed data is likewise beneficial to the FM variants.

show abstract

Fight Fire with Fire: Towards Robust Recommender Systems via Adversarial Poisoning Training

Cited by 26 publications

References 37 publications

Mitigating Human and Computer Opinion Fraud via Contrastive Learning

Mitigating Human and Computer Opinion Fraud via Contrastive Learning

Factorized Structured Regression for Large-Scale Varying Coefficient Models

Boosting Factorization Machines via Saliency-Guided Mixup

Contact Info

Product

Resources

About