Opening the Black Box: Interpretable Remedies for Popularity Bias in Recommender Systems
By: Parviz Ahmadov, Masoud Mansoury
Potential Business Impact:
Makes sure everyone's favorite things get shown.
Popularity bias is a well-known challenge in recommender systems, where a small number of popular items receive disproportionate attention, while the majority of less popular items are largely overlooked. This imbalance often results in reduced recommendation quality and unfair exposure of items. Although existing mitigation techniques address this bias to some extent, they typically lack transparency in how they operate. In this paper, we propose a post-hoc method using a Sparse Autoencoder (SAE) to interpret and mitigate popularity bias in deep recommendation models. The SAE is trained to replicate a pre-trained model's behavior while enabling neuron-level interpretability. By introducing synthetic users with clear preferences for either popular or unpopular items, we identify neurons encoding popularity signals based on their activation patterns. We then adjust the activations of the most biased neurons to steer recommendations toward fairer exposure. Experiments on two public datasets using a sequential recommendation model show that our method significantly improves fairness with minimal impact on accuracy. Moreover, it offers interpretability and fine-grained control over the fairness-accuracy trade-off.
Similar Papers
Investigating Popularity Bias Amplification in Recommender Systems Employed in the Entertainment Domain
Information Retrieval
Fixes unfair movie and music suggestions.
Sparse Autoencoders for Sequential Recommendation Models: Interpretation and Flexible Control
Information Retrieval
Explains why computers suggest what they do.
Extracting Interaction-Aware Monosemantic Concepts in Recommender Systems
Information Retrieval
Makes movie suggestions better by understanding what you like.