Mitigating Popularity Bias in Recommendation with Unbalanced Interactions: A Gradient Perspective
Authors:
- 1 November 2022
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 438-447
Abstract
Recommender systems learn from historical user-item interactions to identify preferred items for target users. These observed interactions are usually unbalanced following a long-tailed distribution. Such long-tailed data lead to popularity bias to recommend popular but not personalized items to users. We present a gradient perspective to understand two negative impacts of popularity bias in recommendation model optimization: (i) the gradient direction of popular item embeddings is closer to that of positive interactions, and (ii) the magnitude of positive gradient for popular items are much greater than that of unpopular items. To address these issues, we propose a simple yet efficient framework to mitigate popularity bias from a gradient perspective. Specifically, we first normalize each user embedding and record accumulated gradients of users and items via popularity bias measures in model training. To address the popularity bias issues, we develop a gradient-based embedding adjustment approach used in model testing. This strategy is generic, model-agnostic, and can be seamlessly integrated into most existing recommender systems. Our extensive experiments on two classic recommendation models and four real-world datasets demonstrate the effectiveness of our method over state-of-the-art debiasing baselines.Keywords
All Related Versions
Funding Information
- National Science Foundation
This publication has 40 references indexed in Scilit:
- The Adressa dataset for news recommendation2017
- Deep Matrix Factorization Models for Recommender Systems2017
- Robust Mapping Learning for Multi-view Multi-label Classification with Missing Labels2017
- Unbiased Learning-to-Rank with Biased Feedback2017
- Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studiesStatistics in Medicine, 2015
- Improving pairwise learning for item recommendation from implicit feedback2014
- My head is your tail2011
- Friendship and mobility2011
- Personalized news recommendation based on click behavior2010
- Causality2009