Recommender Transformers with Behavior Pathways

TL;DR

This paper introduces Recommender Transformers with a novel Pathway Attention mechanism that dynamically identifies and activates pivotal user behavior pathways, significantly improving sequential recommendation accuracy.

Contribution

It proposes a new Pathway Attention mechanism for transformers that selectively focuses on key user behaviors, enhancing recommendation performance.

Findings

RETR achieves state-of-the-art results on seven real-world datasets.

The Pathway Attention mechanism effectively filters trivial behaviors.

RETR outperforms existing sequential recommendation models.

Abstract

Sequential recommendation requires the recommender to capture the evolving behavior characteristics from logged user behavior data for accurate recommendations. However, user behavior sequences are viewed as a script with multiple ongoing threads intertwined. We find that only a small set of pivotal behaviors can be evolved into the user's future action. As a result, the future behavior of the user is hard to predict. We conclude this characteristic for sequential behaviors of each user as the Behavior Pathway. Different users have their unique behavior pathways. Among existing sequential models, transformers have shown great capacity in capturing global-dependent characteristics. However, these models mainly provide a dense distribution over all previous behaviors using the self-attention mechanism, making the final predictions overwhelmed by the trivial behaviors not adjusted to each user. In this paper, we build the Recommender Transformer (RETR) with a novel Pathway Attention mechanism. RETR can dynamically plan the behavior pathway specified for each user, and sparingly activate the network through this behavior pathway to effectively capture evolving patterns useful for recommendation. The key design is a learned binary route to prevent the behavior pathway from being overwhelmed by trivial behaviors. We empirically verify the effectiveness of RETR on seven real-world datasets and RETR yields state-of-the-art performance.