Press release on the AI of waves

https://www.uniroma1.it/it/notizia/lintelligenza-delle-onde-1

https://www.cnr.it/en/press-release/9652/l-intelligenza-delle-onde

https://www.phys.uniroma1.it/fisica/archivionotizie/theory-neuromorphic-computing-waves

https://www.virgilio.it/italia/enna/notizielocali/l_intelligenza_delle_onde-63443731.html

https://www.virgilio.it/italia/enna/notizielocali/l_intelligenza_delle_onde-63443731.html

https://it.finance.yahoo.com/notizie/da-cnr-modello-intelligenza-artificiale-che-sfrutta-onde-124603891.html

https://www.e-gazette.it/sezione/tecnologia/incredibol-sapienza-cnr-studiano-intelligenza-onde

Electomagazine.it

https://ilcorrieredelweb.blogspot.com/2020/09/scienza-cnr-lintelligenza-delle-onde.html

https://www.lescienze.it/news/2020/09/22/news/l_intelligenza_delle_onde-4800586/

https://it.notizie.yahoo.com/da-cnr-modello-intelligenza-artificiale-che-sfrutta-onde-124603891.html

Nuove Direzioni numero 62, Nov-Dic 2020, pag. 75 “L’intelligenza delle onde”

Messaggero, 5 October 2020

Interview of Claudio Conti (podcast)

https://rbe.it/2020/10/22/allenare-le-intelligenze-artificiali-con-le-onde/

Our “machine learning with nonlinear waves” paper featured in Physics!

Riding waves in Neuromorphic Computing, Marios Mattheakis highlights with a thoughtful viewpoint our recent paper in PRL on the artificial intelligence of nonlinear waves.

The Artificial Intelligence of Waves

In a paper published in Physical Review Letters, with title

Theory of Neuromorphic Computing by Waves: Machine Learning by Rogue Waves, Dispersive Shocks and Solitons

we study artificial neural networks with nonlinear waves as a computing reservoir. We discuss universality and the conditions to learn a dataset in terms of output channels and nonlinearity. A feed-forward three-layered model, with an encoding input layer, a wave layer, and a decoding readout, behaves as a conventional neural network in approximating mathematical functions, real-world datasets, and universal Boolean gates.
The rank of the transmission matrix has a fundamental role in assessing the learning abilities of the wave.
For a given set of training points, a threshold nonlinearity for universal interpolation exists. When considering the nonlinear Schrödinger equation, the use of highly nonlinear regimes implies that solitons, rogue, and shock waves do have a leading role in training and computing. Our results may enable the realization of novel machine learning devices by using diverse physical systems, as nonlinear optics, hydrodynamics, polaritonics, and Bose-Einstein condensates. The application of these concepts to photonics opens the way to
a large class of accelerators and new computational paradigms. In complex wave systems, as multimodal fibers, integrated optical circuits, random, topological devices, and metasurfaces, nonlinear waves can be employed to perform computation and solve complex combinatorial optimization.

The paper was selected as Editors’Suggestion and Featured in Physics

See also

https://arxiv.org/abs/1912.07044

Theory of neuromorphic computing by waves

Machine-learning by rogue waves, dispersive shocks, and solitons

We study artificial neural networks with nonlinear waves as a computing reservoir. We discuss universality and the conditions to learn a dataset in terms of output channels and nonlinearity. A feed-forward three-layer model, with an encoding input layer, a wave layer, and a decoding readout, behaves as a conventional neural network in approximating mathematical functions, real-world datasets, and universal Boolean gates. The rank of the transmission matrix has a fundamental role in assessing the learning abilities of the wave. For a given set of training points, a threshold nonlinearity for universal interpolation exists. When considering the nonlinear Schroedinger equation, the use of highly nonlinear regimes implies that solitons, rogue, and shock waves do have a leading role in training and computing. Our results may enable the realization of novel machine learning devices by using diverse physical systems, as nonlinear optics, hydrodynamics, polaritonics, and Bose-Einstein condensates. The application of these concepts to photonics opens the way to a large class of accelerators and new computational paradigms. In complex wave systems, as multimodal fibers, integrated optical circuits, random, topological devices, and metasurfaces, nonlinear waves can be employed to perform computation and solve complex combinatorial optimization.

ArXiv:1912.07077