# Dynamic Modeling of User Preferences for Stable Recommendations

arXiv: https://arxiv.org/abs/2104.05047 · HTML: https://mlsanity.com/paper/2104.05047/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 16 references that are in the map.
- First posted on arXiv 11 Apr 2021.

## Builds on (most cited first, 4 of 4)

- [Image-based Recommendations on Styles and Substitutes](https://mlsanity.com/paper/1506.04757/) (2015)
- [HybridSVD: When Collaborative Information is Not Enough](https://mlsanity.com/paper/1802.06398/) (2018)
- [EigenRec: Generalizing PureSVD for Effective and Efficient Top-N Recommendations](https://mlsanity.com/paper/1511.06033/) (2015)
- [A new simple and effective measure for bag-of-word inter-document similarity measurement](https://mlsanity.com/paper/1902.03402/) (2019)

## Built on by (most cited first, 4 of 4)

- [Recommendation Is a Dish Better Served Warm](https://mlsanity.com/paper/2508.07856/) (2025)
- [Scalable LinUCB: Low-Rank Design Matrix Updates for Recommenders with Large Action Spaces](https://mlsanity.com/paper/2510.19349/) (2025)
- [Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layers](https://mlsanity.com/paper/2509.15113/) (2025)
- [Dynamic Collaborative Filtering for Matrix- and Tensor-based Recommender Systems](https://mlsanity.com/paper/2312.10064/) (2023)

## Abstract (arXiv)

In domains where users tend to develop long-term preferences that do not change too frequently, the stability of recommendations is an important factor of the perceived quality of a recommender system. In such cases, unstable recommendations may lead to poor personalization experience and distrust, driving users away from a recommendation service. We propose an incremental learning scheme that mitigates such problems through the dynamic modeling approach. It incorporates a generalized matrix form of a partial differential equation integrator that yields a dynamic low-rank approximation of time-dependent matrices representing user preferences. The scheme allows extending the famous PureSVD approach to time-aware settings and significantly improves its stability without sacrificing the accuracy in standard top-n recommendations tasks.