A Deep Learning Blueprint for Relational Databases
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F23%3A00373584" target="_blank" >RIV/68407700:21230/23:00373584 - isvavai.cz</a>
Result on the web
<a href="https://nips.cc/virtual/2023/81289" target="_blank" >https://nips.cc/virtual/2023/81289</a>
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
A Deep Learning Blueprint for Relational Databases
Original language description
We introduce a modular neural message-passing scheme that closely follows the formal model of relational databases, effectively enabling end-to-end deep learning directly from database storages. We experiment with several instantiations of the scheme, including notably the use of cross-attention modules to capture the referential constraints of the relational model. We address the issues of efficient learning data representation and loading, salient to the database setting, and compare against representative models from a number of related fields, demonstrating favorable initial results.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
R - Projekt Ramcoveho programu EK
Others
Publication year
2023
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Table Representation Learning Workshop @ NeurIPS
ISBN
—
ISSN
1049-5258
e-ISSN
1049-5258
Number of pages
11
Pages from-to
—
Publisher name
OpenReview.net / University of Massachusetts
Place of publication
Massachusetts
Event location
New Orleans
Event date
Dec 10, 2023
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—