Sum Epsilon-Tube Error Fitness Function Design for GP Symbolic Regression: Preliminary Study

Matoušek, Radomil; Hulka, Tomas; Dobrovský, Ladislav; Kůdela, Jakub

doi:10.1109/iccairo47923.2019.00021

Search citation statements

Order By: Relevance

Paper Sections

Select...

Related Work1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2021

2022

Publication Types

Select...

Other1

Book1

Article1

Relationship

Self Cite0

Independent3

Authors

Journals

Cited by 3 publications

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Staats et al [6] demonstrated the benefits of using TensorFlow to vectorized GP fitness data in both CPU and GPU architectures, achieving performance increases of up to 875 fold for certain classification problems. The engine that the authors developed, KarooGP, is still used to tackle many symbolic regression and classification problems [21][22][23]. However, KarooGP does not take advantage of recent additions to TensorFlow execution models.…”

Section: Related Workmentioning

confidence: 99%

TensorGP -- Genetic Programming Engine in TensorFlow

Baeta¹,

Correia²,

Martins³

et al. 2021

Preprint

View full text Add to dashboard Cite

In this paper, we resort to the TensorFlow framework to investigate the benefits of applying data vectorization and fitness caching methods to domain evaluation in Genetic Programming. For this purpose, an independent engine was developed, TensorGP, along with a testing suite to extract comparative timing results across different architectures and amongst both iterative and vectorized approaches. Our performance benchmarks demonstrate that by exploiting the TensorFlow eager execution model, performance gains of up to two orders of magnitude can be achieved on a parallel approach running on dedicated hardware when compared to a standard iterative approach.

show abstract