Is Contrastive Learning Necessary? A Study of Data Augmentation vs Contrastive Learning in Sequential Recommendation
Keywords: Data Augmentation, Sequential Recommendation, Contrastive Learning
Abstract: Sequential recommender systems (SRS) are designed to predict users’ future behaviors based on their historical interaction data. Recent research has increasingly utilized contrastive learning (CL) to leverage unsupervised signals to alleviate the data sparsity issue in SRS. In general, CL-based SRS first augments the raw sequential interaction data by using data augmentation strategies and employs a contrastive training scheme to enforce the representations of those sequences from the same raw interaction data to be similar. Despite the growing popularity of CL, data augmentation, as a basic component of CL, has not received sufficient attention. This raises the question: Is data augmentation sufficient to achieve superior recommendation results? To answer this question, we benchmark a large amount of data augmentation strategies, as well as state-of-the-art CL-based SRS methods, on four real-world datasets under both warm- and cold-start settings. Intriguingly, the conclusion drawn from our study is that data augmentation is sufficient and CL may not be necessarily required. In fact, utilizing augmentation alone can significantly alleviate the data sparsity issue and certain data augmentation can achieve similar or even superior performance compared with CL-based methods. We hope that our study can further inspire more fundamental studies on the key functional components of complex CL techniques. Our processed datasets and codes will be released once our paper is accepted.
Track: User Modeling and Recommendation
Submission Guidelines Scope: Yes
Submission Guidelines Blind: Yes
Submission Guidelines Format: Yes
Submission Guidelines Limit: Yes
Submission Guidelines Authorship: Yes
Student Author: Yes
Submission Number: 2080
Loading