Differential Evolution Algorithm for Compact and Highly Accurate Photonic Extreme Learning Machines
We employ a differential evolution algorithm to reduce the impact of the random layer in programmable photonic extreme learning machines. Higher accuracy with reduced node number requirements is demonstrated in an integrated hexagonal mesh.
Authors
- Daniel Pérez (ORCID: https://orcid.org/0000-0002-0873-205X)
- Antonio Hurtado (ORCID: https://orcid.org/0000-0002-4448-9034)
- J. Capmany (ORCID: https://orcid.org/0000-0002-6460-4167)
- José Roberto Rausell‐Campo (ORCID: https://orcid.org/0000-0002-6710-4766)
Institutions
- University of Strathclyde (GB)
- iPronics Programmable Photonics (Spain) (ES)
Publication Details
- Published
- 2024-08-04
- DOI
- https://doi.org/10.1109/cleo-pr60912.2024.10676608
- Primary Topic
- Machine Learning and ELM
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Engineering and Physical Sciences Research Council
- H2020 European Research Council