Multi-Task Learning For Reduced Popularity Bias In Multi-Territory Video Recommendations

TL;DR

This paper introduces a multi-task learning framework with adaptive upsampling to mitigate popularity bias in multi-territory video recommendations, significantly improving recommendation accuracy across regions.

Contribution

The paper presents a novel multi-task learning approach combined with adaptive upsampling to reduce popularity bias and learn geographic user embeddings in multi-territory recommender systems.

Findings

Up to 65.27% relative gain in PR-AUC metric.

Effective reduction of popularity bias for global items.

Improved recommendation performance across multiple territories.

Abstract

Various data imbalances that naturally arise in a multi-territory personalized recommender system can lead to a significant item bias for globally prevalent items. A locally popular item can be overshadowed by a globally prevalent item. Moreover, users' viewership patterns/statistics can drastically change from one geographic location to another which may suggest to learn specific user embeddings. In this paper, we propose a multi-task learning (MTL) technique, along with an adaptive upsampling method to reduce popularity bias in multi-territory recommendations. Our proposed framework is designed to enrich training examples with active users representation through upsampling, and capable of learning geographic-based user embeddings by leveraging MTL. Through experiments, we demonstrate the effectiveness of our framework in multiple territories compared to a baseline not incorporating our proposed techniques.~Noticeably, we show improved relative gain of up to $65.27\%$ in PR-AUC metric. A case study is presented to demonstrate the advantages of our methods in attenuating the popularity bias of global items.