Abstract
We emulate and experimentally validate the scaling of SOA-based all-optical deep neural networks by accurately modeling the OSNR degradation in a chain of SOAs: The photonic neural network can scale up to 16 layers when using only 4 neurons/layer without notable accuracy degradation.
| Original language | English |
|---|---|
| Title of host publication | 2021 Asia Communications and Photonics Conference, ACP 2021 - Proceedings |
| Publisher | Optical Society of America (OSA) |
| Pages | T1E.4 |
| Number of pages | 3 |
| ISBN (Electronic) | 9781957171005 |
| ISBN (Print) | 978-1-957171-00-5 |
| DOIs | |
| Publication status | Published - 2021 |
| Event | 2021 Asia Communications and Photonics Conference, ACP 2021 - Shanghai, China Duration: 24 Oct 2021 → 27 Oct 2021 https://www.osapublishing.org/conference.cfm?meetingid=144&yr=2021 |
Conference
| Conference | 2021 Asia Communications and Photonics Conference, ACP 2021 |
|---|---|
| Country/Territory | China |
| City | Shanghai |
| Period | 24/10/21 → 27/10/21 |
| Internet address |
Bibliographical note
Publisher Copyright:© Optica Publishing Group 2021
Keywords
- All-optical Neural Network
- Semiconductor optical amplifiers
- Optical networks
- Optical signal to noise ratio
- Optical systems
- Spontaneous emission