Contrastive Learning for Diversity-Aware Product Recommendations in Retail

2026-02-09

AI Summary

This work addresses the insufficient coverage of long-tail items and lack of diversity in recommendation systems caused by the dominance of popular items. To mitigate popularity bias, the authors integrate contrastive learning into IKEA’s large-scale retail recommendation system and devise a tailored negative sampling strategy. The proposed approach significantly enhances catalog coverage and recommendation diversity while preserving recommendation accuracy. Both offline and online experiments demonstrate that the method effectively balances diversity and performance, offering a practical pathway toward fairness and exploratory optimization in industrial-scale recommender systems.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Machine Learning: Learning Preferences or Rankings
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Economics and fairness of platforms and recommendation systems

Abstract

Recommender systems often struggle with long-tail distributions and limited item catalog exposure, where a small subset of popular items dominates recommendations. This challenge is especially critical in large-scale online retail settings with extensive and diverse product assortments. This paper introduces an approach to enhance catalog coverage without compromising recommendation quality in the existing digital recommendation pipeline at IKEA Retail. Drawing inspiration from recent advances in negative sampling to address popularity bias, we integrate contrastive learning with carefully selected negative samples. Through offline and online evaluations, we demonstrate that our method improves catalog coverage, ensuring a more diverse set of recommendations yet preserving strong recommendation performance.

Problem

• contrastive learning
• diversity-aware recommendations
• popularity bias
• catalog coverage
• long-tail distribution

Innovation

• contrastive learning
• negative sampling
• popularity bias
• catalog coverage
• diverse recommendations