Linear combinations of features as leaf nodes in symbolic regression

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F17%3A00313865" target="_blank" >RIV/68407700:21230/17:00313865 - isvavai.cz</a>
Alternative codes found
RIV/68407700:21730/17:00313865
Result on the web
<a href="https://dl.acm.org/citation.cfm?id=3076009" target="_blank" >https://dl.acm.org/citation.cfm?id=3076009</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1145/3067695.3076009" target="_blank" >10.1145/3067695.3076009</a>

Result language
angličtina
Original language name
Linear combinations of features as leaf nodes in symbolic regression
Original language description
We propose a new type of leaf node for use in Symbolic Regression (SR) that performs linear combinations of feature variables (LCF). LCF's weights are tuned using a gradient method based on back-propagation algorithm known from neural networks. Multi-Gene Genetic Programming (MGGP) was chosen as a baseline model. As a sanity check, we experimentally show that LCFs improve the performance of the baseline on a rotated toy SR problem. We then perform a thorougher experimental study on a number of artificial and real-world SR benchmarks. The usage of LCFs in MGGP statically improved the results in 5 cases out of 9, while it worsen them in only a single case.
Czech name
—
Czech description
—

Project
<a href="/en/project/GA15-22731S" target="_blank" >GA15-22731S: Symbolic Regression for Reinforcement Learning in Continuous Spaces</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2017
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Proceedings of the Genetic and Evolutionary Computation Conference Companion
ISBN
978-1-4503-4939-0
ISSN
—
e-ISSN
—
Number of pages
2
Pages from-to
145-146
Publisher name
ACM
Place of publication
New York
Event location
Berlín
Event date
Jul 15, 2017
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—

Similar results(10)