# Modeling Churn in Recommender Systems with Aggregated Preferences

arXiv: https://arxiv.org/abs/2502.18483 · HTML: https://mlsanity.com/paper/2502.18483/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 48 references that are in the map.
- First posted on arXiv 9 Feb 2025.

## Builds on (most cited first, 16 of 16)

- [Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling](https://mlsanity.com/paper/1412.3555/) (2014)
- [Meta Matrix Factorization for Federated Rating Predictions](https://mlsanity.com/paper/1910.10086/) (2019)
- [Learning to branch with Tree MDPs](https://mlsanity.com/paper/2205.11107/) (2022)
- [Task-adaptive Neural Process for User Cold-Start Recommendation](https://mlsanity.com/paper/2103.06137/) (2021)
- [Transfer-Meta Framework for Cross-domain Recommendation to Cold-Start Users](https://mlsanity.com/paper/2105.04785/) (2021)
- [Reinforcement Learning for Branch-and-Bound Optimisation using Retrospective Trajectories](https://mlsanity.com/paper/2205.14345/) (2022)
- [Personalized Adaptive Meta Learning for Cold-start User Preference Prediction](https://mlsanity.com/paper/2012.11842/) (2020)
- [A Heterogeneous Information Network based Cross Domain Insurance Recommendation System for Cold Start Users](https://mlsanity.com/paper/2007.15293/) (2020)
- [Optimal Recommendation to Users that React: Online Learning for a Class of POMDPs](https://mlsanity.com/paper/1603.09233/) (2016)
- [Partially Observable Markov Decision Process for Recommender Systems](https://mlsanity.com/paper/1608.07793/) (2016)
- [Modeling Attrition in Recommender Systems with Departing Bandits](https://mlsanity.com/paper/2203.13423/) (2022)
- [Fatigue-aware Bandits for Dependent Click Models](https://mlsanity.com/paper/2008.09733/) (2020)
- [Goodbye Tracking? Impact of iOS App Tracking Transparency and Privacy Labels](https://mlsanity.com/paper/2204.03556/) (2022)
- [Maximizing Cumulative User Engagement in Sequential Recommendation: An Online Optimization Perspective](https://mlsanity.com/paper/2006.04520/) (2020)
- [Interactive and Explainable Point-of-Interest Recommendation using Look-alike Groups](https://mlsanity.com/paper/2009.00099/) (2020)
- [Anonymous Learning via Look-Alike Clustering: A Precise Analysis of Model Generalization](https://mlsanity.com/paper/2310.04015/) (2023)

## Built on by (most cited first, 1 of 1)

- [When to Switch, Not Just What: Transition Quality Prediction in Clash Royale](https://mlsanity.com/paper/2605.21868/) (2026)

## Abstract (arXiv)

While recommender systems (RSs) traditionally rely on extensive individual user data, regulatory and technological shifts necessitate reliance on aggregated user information. This shift significantly impacts the recommendation process, requiring RSs to engage in intensive exploration to identify user preferences. However, this approach risks user churn due to potentially unsatisfactory recommendations. In this paper, we propose a model that addresses the dual challenges of leveraging aggregated user information and mitigating churn risk. Our model assumes that the RS operates with a probabilistic prior over user types and aggregated satisfaction levels for various content types. We demonstrate that optimal policies naturally transition from exploration to exploitation in finite time, develop a branch-and-bound algorithm for computing these policies, and empirically validate its effectiveness.