Lightweight method improves forgetting in sequential recommender systems
SURF: Subtractive Updates for Recommender Forgetting
Information Retrieval
Summary
Recommender systems show you what you might like based on your past behavior, but sometimes they need to "forget" some of that data, for example to respect privacy laws. The authors designed a new way to help these systems forget specific items without needing to completely retrain the model, which usually takes a lot of time. Their method, SURF, smartly updates only a small part related to the item to forget. Tests showed this approach is almost as accurate as full retraining but much faster.
What this means in practice
- •For mobile app developers: Implement faster privacy-compliant data removal in apps using sequential recommendation to reduce retraining time drastically.
- •For e-commerce platform engineers: Offer customer data deletion that respects sequential user data without full model retraining, saving computational resources.
Authors
Filippo Betello, Antonio Purificato, Nicola Tonellotto, Fabrizio Silvestri
Abstract
The increasing demand for user privacy and compliance with regulations such as GDPR has made machine unlearning a fundamental requirement for modern recommender systems. However, Sequential Recommender Systems (SRS) pose unique challenges for unlearning due to their reliance on temporal interaction patterns. Existing approaches either require computationally prohibitive full retraining or fail to account for the sequential nature of user behavior. We propose SURF (Subtractive Updates for Recommender Forgetting), a lightweight framework for approximate machine unlearning in SRS. SURF operates in three stages: (i) identifying the neighborhood of the item to forget in the embedding space, (ii) training an auxiliary model on this compact local subset, and (iii) subtracting the auxiliary model's scores from the original model at inference time. Experiments against five baselines on 7 datasets show that SURF achieves unlearning effectiveness comparable to full retraining while substantially reducing computational cost, yielding up to a 32% improvement in NDCG@20 while requiring just 2% of the original retraining baseline time budget. We share our code at https://github.com/FilippoBetello/SURF.