Gumble Softmax For User Behavior Modeling

TL;DR

This paper introduces RDRSR, a sequential recommendation model that dynamically adjusts the number of user interest representations over time using Gumbel-Softmax, bi-directional self-attention, and interest allocation modules.

Contribution

It proposes a novel approach to model evolving user interests with a dynamic number of representations, improving personalization in recommendation systems.

Findings

Outperforms existing models on real-world datasets.

Effectively captures evolving user interests.

Demonstrates the importance of dynamic interest modeling.

Abstract

Recently, sequential recommendation systems are important in solving the information overload in many online services. Current methods in sequential recommendation focus on learning a fixed number of representations for each user at any time, with a single representation or multi representations for the user. However, when a user is exploring items on an e-commerce recommendation system, the number of this user's hobbies may change overtime (e.g. increase/reduce one more interest), affected by the user's evolving self needs. Moreover, different users may have various number of interests. In this paper, we argue that it is meaningful to explore a personalized dynamic number of user interests, and learn a dynamic group of user interest representations accordingly. We propose a sequential model with dynamic number of representations for recommendation systems (RDRSR). Specifically, RDRSR is composed of a dynamic interest discriminator (DID) module and a dynamic interest allocator (DIA) module. The DID module explores the number of a user's interests by learning the overall sequential characteristics with bi-directional self-attention and Gumbel-Softmax. The DIA module make the historical clicked items into a group of item groups and constructs user's dynamic interest representation. Additionally, experiments on the real-world datasets demonstrates our model's effectiveness.