The optical reservoir computer: a new approach to a programmable integrated optics system based on an artificial neural network

Phang, Sendy; Sewell, Phillip D.; Vukovic, Ana; Benson, Trevor M.

Abstract

In this chapter, we briefly reviewed some developments in the field of integrated optics since Miller first introduced this concept, noting some approaches to optimise structures based on functional performance criteria. Chapter Contents: • 12.1 Introduction • 12.2 Some applications of genetic algorithms in integrated optics design • 12.3 Functional integrated optics powered by a reservoir computer • 12.3.1 Introduction to an algorithmic reservoir computer • 12.3.2 An optical reservoir computer as a temporal signal discriminator • 12.3.3 Chaotic cavity as a reservoir computer kernel • 12.3.4 ORC training and validation • 12.4 Conclusions • References.

Citation

Phang, S., Sewell, P. D., Vukovic, A., & Benson, T. M. (2020). The optical reservoir computer: a new approach to a programmable integrated optics system based on an artificial neural network. In G. C. Righini, & M. Ferrari (Eds.), Integrated Optics Volume 2: Characterization, devices and applications (361-380). Institution of Engineering and Technology (IET). https://doi.org/10.1049/pbcs077g_ch12

• Acceptance Date: Dec 31, 2020
• Online Publication Date: Dec 31, 2020
• Publication Date: Dec 31, 2020
• Deposit Date: May 18, 2022
• Publisher: Institution of Engineering and Technology (IET)
• Pages: 361-380
• Book Title: Integrated Optics Volume 2: Characterization, devices and applications
• Chapter Number: 12
• ISBN: 9781839533433
• DOI: https://doi.org/10.1049/pbcs077g_ch12
• Public URL: https://nottingham-repository.worktribe.com/output/8132533
• Publisher URL: https://digital-library.theiet.org/content/books/10.1049/pbcs077g_ch12