Noise effects on time delay reservoir computing using silicon microring resonators
Time-delay Reservoir Computing (TRC) are neural networks which thanks to their hardware simplicity are suited for photonic implementation. Here, we numerically investigated and experimentally tested a TRC based on a nonlinear single silicon microring resonator (MRR) coupled to an optical fibre loop...
Saved in:
Main Authors | , , , , |
---|---|
Format | Conference Proceeding |
Language | English |
Published |
SPIE
05.03.2022
|
Online Access | Get full text |
ISBN | 1510648798 9781510648791 |
ISSN | 0277-786X |
DOI | 10.1117/12.2609056 |
Cover
Loading…
Summary: | Time-delay Reservoir Computing (TRC) are neural networks which thanks to their hardware simplicity are suited for photonic implementation. Here, we numerically investigated and experimentally tested a TRC based on a nonlinear single silicon microring resonator (MRR) coupled to an optical fibre loop which provides an external tuneable optical feedback. The work is inspired by the key role that MRRs demonstrate in integrated optical devices, together with the significant advantage they provide due to their fully passive nonlinearity. Insights about the computational properties of the system are provided as a result of the performance on multiple benchmark tasks. |
---|---|
Bibliography: | Conference Date: 2022-01-22|2022-02-28 Conference Location: San Francisco, California, United States |
ISBN: | 1510648798 9781510648791 |
ISSN: | 0277-786X |
DOI: | 10.1117/12.2609056 |