Saved in:
Bibliographic Details
Main Authors: Mazzetto, Alessio, Khalili, Mohammad Mahdi, Nern, Laura Fee, Viderman, Michael, Shtoff, Alex, Dembczyński, Krzysztof
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.15229
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912909372686336
author Mazzetto, Alessio
Khalili, Mohammad Mahdi
Nern, Laura Fee
Viderman, Michael
Shtoff, Alex
Dembczyński, Krzysztof
author_facet Mazzetto, Alessio
Khalili, Mohammad Mahdi
Nern, Laura Fee
Viderman, Michael
Shtoff, Alex
Dembczyński, Krzysztof
contents We address prediction problems on tabular categorical data, where each instance is defined by multiple categorical attributes, each taking values from a finite set. These attributes are often referred to as fields, and their categorical values as features. Such problems frequently arise in practical applications, including click-through rate prediction and social sciences. We introduce and analyze {tensorFM}, a new model that efficiently captures high-order interactions between attributes via a low-rank tensor approximation representing the strength of these interactions. Our model generalizes field-weighted factorization machines. Empirically, tensorFM demonstrates competitive performance with state-of-the-art methods. Additionally, its low latency makes it well-suited for time-sensitive applications, such as online advertising.
format Preprint
id arxiv_https___arxiv_org_abs_2602_15229
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle tensorFM: Low-Rank Approximations of Cross-Order Feature Interactions
Mazzetto, Alessio
Khalili, Mohammad Mahdi
Nern, Laura Fee
Viderman, Michael
Shtoff, Alex
Dembczyński, Krzysztof
Machine Learning
Information Retrieval
68T05, 15A69
I.2.6
We address prediction problems on tabular categorical data, where each instance is defined by multiple categorical attributes, each taking values from a finite set. These attributes are often referred to as fields, and their categorical values as features. Such problems frequently arise in practical applications, including click-through rate prediction and social sciences. We introduce and analyze {tensorFM}, a new model that efficiently captures high-order interactions between attributes via a low-rank tensor approximation representing the strength of these interactions. Our model generalizes field-weighted factorization machines. Empirically, tensorFM demonstrates competitive performance with state-of-the-art methods. Additionally, its low latency makes it well-suited for time-sensitive applications, such as online advertising.
title tensorFM: Low-Rank Approximations of Cross-Order Feature Interactions
topic Machine Learning
Information Retrieval
68T05, 15A69
I.2.6
url https://arxiv.org/abs/2602.15229