Reservoir computing is an unconventional computing paradigm that concerns computing with dynamical systems. Physical systems excited by an external input can do computation, which can then be learned by an output classifier to perform some tasks such as time series prediction. Recent progress in reservoir computing has led to numerous implementations, particularly in photonics. Photonic circuits offer opportunities for dense component integration, scalability, and the miniaturization of existing fiber-based schemes. In this talk, we examine how a photonic reservoir computing system with minimalist hardware requirements—both in the reservoir and the peripheral electronics—performs on standard benchmark tasks.
This Talk will be broadcasted in the following zoom link: https://us06web.zoom.us/j/89027654460?pwd=Wg9TYMPqqP2ipfj2JVvEagmzaTw29c.1
Detalls de contacte:
Miguel C. Soriano Contact form