DiffuRec: A Diffusion Model for Sequential Recommendation

TL;DR

DiffuRec introduces a diffusion-based approach to sequential recommendation, representing items as distributions to better capture user preferences and item aspects, leading to superior performance over existing methods.

Contribution

This paper pioneers the application of diffusion models to sequential recommendation, enabling dynamic item representation generation and uncertainty modeling.

Findings

DiffuRec outperforms strong baselines on four datasets.

Representing items as distributions improves recommendation diversity.

DiffuRec effectively captures user interests and item aspects.

Abstract

Mainstream solutions to Sequential Recommendation (SR) represent items with fixed vectors. These vectors have limited capability in capturing items' latent aspects and users' diverse preferences. As a new generative paradigm, Diffusion models have achieved excellent performance in areas like computer vision and natural language processing. To our understanding, its unique merit in representation generation well fits the problem setting of sequential recommendation. In this paper, we make the very first attempt to adapt Diffusion model to SR and propose DiffuRec, for item representation construction and uncertainty injection. Rather than modeling item representations as fixed vectors, we represent them as distributions in DiffuRec, which reflect user's multiple interests and item's various aspects adaptively. In diffusion phase, DiffuRec corrupts the target item embedding into a Gaussian distribution via noise adding, which is further applied for sequential item distribution representation generation and uncertainty injection. Afterward, the item representation is fed into an Approximator for target item representation reconstruction. In reverse phase, based on user's historical interaction behaviors, we reverse a Gaussian noise into the target item representation, then apply a rounding operation for target item prediction. Experiments over four datasets show that DiffuRec outperforms strong baselines by a large margin.