Fisher-Weighted Merge of Contrastive Learning Models in Sequential Recommendation

TL;DR

This paper introduces a Fisher-Weighted Merge approach for contrastive learning models in sequential recommendation, enhancing model robustness and performance by effectively merging multiple models to better capture evolving user preferences.

Contribution

It is the first to apply Fisher-Merging to sequential recommendation, addressing practical challenges and improving model fine-tuning and overall recommendation accuracy.

Findings

Fisher-Merging improves recommendation performance.

The method effectively handles data sparsity in sequential data.

Enhanced robustness in model fine-tuning.

Abstract

Along with the exponential growth of online platforms and services, recommendation systems have become essential for identifying relevant items based on user preferences. The domain of sequential recommendation aims to capture evolving user preferences over time. To address dynamic preference, various contrastive learning methods have been proposed to target data sparsity, a challenge in recommendation systems due to the limited user-item interactions. In this paper, we are the first to apply the Fisher-Merging method to Sequential Recommendation, addressing and resolving practical challenges associated with it. This approach ensures robust fine-tuning by merging the parameters of multiple models, resulting in improved overall performance. Through extensive experiments, we demonstrate the effectiveness of our proposed methods, highlighting their potential to advance the state-of-the-art in sequential learning and recommendation systems.