Tabular Transformers Meet Relational Databases

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F25%3A00384310" target="_blank" >RIV/68407700:21230/25:00384310 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1145/3749991" target="_blank" >https://doi.org/10.1145/3749991</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1145/3749991" target="_blank" >10.1145/3749991</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Tabular Transformers Meet Relational Databases

Popis výsledku v původním jazyce

Transformer models have continuously expanded into all machine learning domains convertible to the underlying sequence-to-sequence representation, including tabular data. However, while ubiquitous, this representation restricts their extension to the more general case of relational databases. In this paper, we introduce a modular neural message-passing scheme that closely adheres to the formal relational model, enabling direct end-to-end learning of tabular transformers from database storage systems. We address the associated challenges of appropriate learning data representation and loading, which are critical in the database setting, and compare our approach against a number of representative models from various related fields across a significantly wide range of datasets. Our results then demonstrate superior performance of this newly proposed class of neural architectures.

Název v anglickém jazyce

Tabular Transformers Meet Relational Databases

Popis výsledku anglicky

Transformer models have continuously expanded into all machine learning domains convertible to the underlying sequence-to-sequence representation, including tabular data. However, while ubiquitous, this representation restricts their extension to the more general case of relational databases. In this paper, we introduce a modular neural message-passing scheme that closely adheres to the formal relational model, enabling direct end-to-end learning of tabular transformers from database storage systems. We address the associated challenges of appropriate learning data representation and loading, which are critical in the database setting, and compare our approach against a number of representative models from various related fields across a significantly wide range of datasets. Our results then demonstrate superior performance of this newly proposed class of neural architectures.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

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

Projekt

<a href="/cs/project/GA24-11664S" target="_blank" >GA24-11664S: Relační posilované učení pro akceleraci vědy</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2025

Kód důvěrnosti údajů

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

Název periodika

ACM Transactions on Intelligent Systems and Technology

ISSN

2157-6904

e-ISSN

2157-6912

Svazek periodika

16

Číslo periodika v rámci svazku

5

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

24

Strana od-do

555-578

Kód UT WoS článku

001606000100001

EID výsledku v databázi Scopus

2-s2.0-105019641424