Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction
Reservoir computing is a relatively recent computational paradigm that originates from a recurrent neural network and is known for its wide range of implementations using different physical technologies. Large reservoirs are very hard to obtain in conventional computers, as both the computation comp...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Physical Society
2020-11-01
|
| Series: | Physical Review X |
| Online Access: | http://doi.org/10.1103/PhysRevX.10.041037 |
| _version_ | 1818430907453800448 |
|---|---|
| author | Mushegh Rafayelyan Jonathan Dong Yongqi Tan Florent Krzakala Sylvain Gigan |
| author_facet | Mushegh Rafayelyan Jonathan Dong Yongqi Tan Florent Krzakala Sylvain Gigan |
| author_sort | Mushegh Rafayelyan |
| collection | DOAJ |
| description | Reservoir computing is a relatively recent computational paradigm that originates from a recurrent neural network and is known for its wide range of implementations using different physical technologies. Large reservoirs are very hard to obtain in conventional computers, as both the computation complexity and memory usage grow quadratically. We propose an optical scheme performing reservoir computing over very large networks potentially being able to host several millions of fully connected photonic nodes thanks to its intrinsic properties of parallelism and scalability. Our experimental studies confirm that, in contrast to conventional computers, the computation time of our optical scheme is only linearly dependent on the number of photonic nodes of the network, which is due to electronic overheads, while the optical part of computation remains fully parallel and independent of the reservoir size. To demonstrate the scalability of our optical scheme, we perform for the first time predictions on large spatiotemporal chaotic datasets obtained from the Kuramoto-Sivashinsky equation using optical reservoirs with up to 50 000 optical nodes. Our results are extremely challenging for conventional von Neumann machines, and they significantly advance the state of the art of unconventional reservoir computing approaches, in general. |
| first_indexed | 2024-12-14T15:40:52Z |
| format | Article |
| id | doaj.art-f2ccc5e3b5054f3cb286dcedeabf9896 |
| institution | Directory Open Access Journal |
| issn | 2160-3308 |
| language | English |
| last_indexed | 2024-12-14T15:40:52Z |
| publishDate | 2020-11-01 |
| publisher | American Physical Society |
| record_format | Article |
| series | Physical Review X |
| spelling | doaj.art-f2ccc5e3b5054f3cb286dcedeabf98962022-12-21T22:55:37ZengAmerican Physical SocietyPhysical Review X2160-33082020-11-0110404103710.1103/PhysRevX.10.041037Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems PredictionMushegh RafayelyanJonathan DongYongqi TanFlorent KrzakalaSylvain GiganReservoir computing is a relatively recent computational paradigm that originates from a recurrent neural network and is known for its wide range of implementations using different physical technologies. Large reservoirs are very hard to obtain in conventional computers, as both the computation complexity and memory usage grow quadratically. We propose an optical scheme performing reservoir computing over very large networks potentially being able to host several millions of fully connected photonic nodes thanks to its intrinsic properties of parallelism and scalability. Our experimental studies confirm that, in contrast to conventional computers, the computation time of our optical scheme is only linearly dependent on the number of photonic nodes of the network, which is due to electronic overheads, while the optical part of computation remains fully parallel and independent of the reservoir size. To demonstrate the scalability of our optical scheme, we perform for the first time predictions on large spatiotemporal chaotic datasets obtained from the Kuramoto-Sivashinsky equation using optical reservoirs with up to 50 000 optical nodes. Our results are extremely challenging for conventional von Neumann machines, and they significantly advance the state of the art of unconventional reservoir computing approaches, in general.http://doi.org/10.1103/PhysRevX.10.041037 |
| spellingShingle | Mushegh Rafayelyan Jonathan Dong Yongqi Tan Florent Krzakala Sylvain Gigan Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction Physical Review X |
| title | Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction |
| title_full | Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction |
| title_fullStr | Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction |
| title_full_unstemmed | Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction |
| title_short | Large-Scale Optical Reservoir Computing for Spatiotemporal Chaotic Systems Prediction |
| title_sort | large scale optical reservoir computing for spatiotemporal chaotic systems prediction |
| url | http://doi.org/10.1103/PhysRevX.10.041037 |
| work_keys_str_mv | AT musheghrafayelyan largescaleopticalreservoircomputingforspatiotemporalchaoticsystemsprediction AT jonathandong largescaleopticalreservoircomputingforspatiotemporalchaoticsystemsprediction AT yongqitan largescaleopticalreservoircomputingforspatiotemporalchaoticsystemsprediction AT florentkrzakala largescaleopticalreservoircomputingforspatiotemporalchaoticsystemsprediction AT sylvaingigan largescaleopticalreservoircomputingforspatiotemporalchaoticsystemsprediction |