Acton, S and Abramowitz, S and Toledo, L and Nitschke, E (2020) Efficiently Coevolving Deep Neural Networks and Data Augmentations, Proceedings of IEEE Symposium Series on Computational Intelligence (IEEE SSCI 2020).
![[thumbnail of 2020-Efficiently Coevolving Deep Neural Networks and Data Augmentations.pdf]](https://pubs.cs.uct.ac.za/style/images/fileicons/text.png) |
Text
2020-Efficiently Coevolving Deep Neural Networks and Data Augmentations.pdf Download (693kB) |
Abstract
Designing large deep learning neural networks by hand requires tuning large sets of method parameters, requiring trial and error testing and domain specific knowledge. Neuroevolution methods such as CoDeepNeat (CDN), based on Neuroevolution of Augmenting Topologies (NEAT), apply evolutionary algorithms to automate deep neural network parameter optimisation. This paper presents and demonstrates various novel beneficial extensions to the CDN method, including new genotypic speciation mechanisms, special mappings in deep neural network encodings, as well as evolving Data Augmentation schemes. Results indicate that these CDN method variants yield significant task-performance benefits over the benchmark CDN method when evaluated on a popular public image recognition data-set.
| Item Type: | Conference paper |
|---|---|
| Subjects: | Computing methodologies > Machine learning |
| Date Deposited: | 03 Dec 2021 11:15 |
| Last Modified: | 03 Dec 2021 11:15 |
| URI: | https://pubs.cs.uct.ac.za/id/eprint/1486 |
Actions (login required)
 |
View Item |