Probabilistic Modeling, Learnability and Uncertainty Estimation for Interaction Prediction in Movie Rating Datasets

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F25%3A00387677" target="_blank" >RIV/68407700:21240/25:00387677 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1145/3705328.3759332" target="_blank" >https://doi.org/10.1145/3705328.3759332</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3705328.3759332" target="_blank" >10.1145/3705328.3759332</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Probabilistic Modeling, Learnability and Uncertainty Estimation for Interaction Prediction in Movie Rating Datasets

Popis výsledku v původním jazyce

In this paper, we examine the hypothesis that the interactions recorded in many Recommendation Systems datasets are distributed according to a low-rank distribution, i.e. a mixture of factorizable distributions. Surprisingly, we find that on several popular datasets, a simple non-negative matrix factorization method equals or outperforms more modern methods such as LightGCN, which indicates that the sampling distribution over interactions is indeed low-rank. Furthermore, we mathematically prove that low-rank distributions are learnable with a sparse number of observations (where m/n and r refer to the number of users/items and the non-negative rank respectively) both in terms of the total variation norm and in terms of the expected recall at k, arguably providing some of the first generalization bounds for recommender systems in the implicit feedback setting. We also provide a modified version of the NMF algorithm which provides further performance improvements compared to the standard NMF baseline on the smaller datasets considered. Finally, we propose the theoretically grounded concept of empirical expected recall as an uncertainty estimate for probabilistic models of the recommendation task, and demonstrate its success in a setting where user-wise abstentions are allowed.

Název v anglickém jazyce

Probabilistic Modeling, Learnability and Uncertainty Estimation for Interaction Prediction in Movie Rating Datasets

Popis výsledku anglicky

In this paper, we examine the hypothesis that the interactions recorded in many Recommendation Systems datasets are distributed according to a low-rank distribution, i.e. a mixture of factorizable distributions. Surprisingly, we find that on several popular datasets, a simple non-negative matrix factorization method equals or outperforms more modern methods such as LightGCN, which indicates that the sampling distribution over interactions is indeed low-rank. Furthermore, we mathematically prove that low-rank distributions are learnable with a sparse number of observations (where m/n and r refer to the number of users/items and the non-negative rank respectively) both in terms of the total variation norm and in terms of the expected recall at k, arguably providing some of the first generalization bounds for recommender systems in the implicit feedback setting. We also provide a modified version of the NMF algorithm which provides further performance improvements compared to the standard NMF baseline on the smaller datasets considered. Finally, we propose the theoretically grounded concept of empirical expected recall as an uncertainty estimate for probabilistic models of the recommendation task, and demonstrate its success in a setting where user-wise abstentions are allowed.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

RecSys '25: Proceedings of the Nineteenth ACM Conference on Recommender Systems

ISBN

979-8-4007-1364-4

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

1261-1266

Název nakladatele

ACM

Místo vydání

New York

Místo konání akce

Prague

Datum konání akce

22. 9. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001572100200183