Parsimonious Optimization of Multitask Neural Network Hyperparameters
Parsimonious Optimization of Multitask Neural Network Hyperparameters
About this item
Full title
Author / Creator
Publisher
Basel: MDPI AG
Journal title
Language
English
Formats
Publication information
Publisher
Basel: MDPI AG
Subjects
More information
Scope and Contents
Contents
Neural networks are rapidly gaining popularity in chemical modeling and Quantitative Structure–Activity Relationship (QSAR) thanks to their ability to handle multitask problems. However, outcomes of neural networks depend on the tuning of several hyperparameters, whose small variations can often strongly affect their performance. Hence, optimizatio...
Alternative Titles
Full title
Parsimonious Optimization of Multitask Neural Network Hyperparameters
Authors, Artists and Contributors
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_doaj_primary_oai_doaj_org_article_6efa002dff22447b824520ca9609aeec
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_doaj_primary_oai_doaj_org_article_6efa002dff22447b824520ca9609aeec
Other Identifiers
ISSN
1420-3049
E-ISSN
1420-3049
DOI
10.3390/molecules26237254
