SQL4NN: Validation and Expressive Querying of Models as Data

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/47403

Title:	SQL4NN: Validation and Expressive Querying of Models as Data
Authors:	Gerarts, Mark STEEGMANS, Juno VAN DEN BUSSCHE, Jan
Issue Date:	2025
Publisher:	ASSOC COMPUTING MACHINERY
Source:	Proceedings of the ninth workshop on data managment for end-to-end machine learning, DEEM, ASSOC COMPUTING MACHINERY, p. 1 -5 (Art N° 10)
Abstract:	We consider machine learning models, learned from data, to be an important, intensional kind of data in themselves. As such, various analysis tasks on models can be thought of as queries over this intensional data, often combined with extensional data such as data for training or validation. We argue that relational database systems and SQL can be well suited for many such tasks.
Notes:	Gerarts, M (corresponding author), Hasselt Univ, Hasselt, Belgium. mark.gerarts@student.uhasselt.be; juno.steegmans@uhasselt.be; jan.vandenbussche@uhasselt.be
Keywords:	neural network;CCS Concepts;Information systems → Query languages;;ReLU;intensional data;Computing methodologies → Machine learning Keywords neural network, ReLU, intensional data, piecewise linear function, integration, pruning, saliency map;piecewise linear function;integration;pruning;saliency map
Document URI:	http://hdl.handle.net/1942/47403
ISBN:	979-8-4007-1924-0
DOI:	10.1145/3735654.3735946
ISI #:	001532062900010
Rights:	2025 Copyright held by the owner/author(s). Publication rights licensed to ACM
Category:	C1
Type:	Proceedings Paper
Appears in Collections:	Research publications

File	Description	Size	Format
SQL4NN_ Validation and Expressive Querying of Models as Data.pdf Restricted Access	Published version	638.52 kB	Adobe PDF	View/Open Request a copy

Check