High Performance Optical Reservoir Computing Based on Spatially Extended Systems

Our work targets efficient (high speed and low power) analogue photonic computing systems based on the concept of reservoir computing (RC) [1,2]. This framework allows to exploit the transient dynamics of a nonlinear dynamical system for performing useful computations. Many photonic implementations have been realized [3], and relevant examples are discussed in Refs [4,5]. In this work we consider a reservoir consisting of a spatially extended non-linear photonic system with diffusive coupling between the different regions. The optical field in these regions represent the internal variables of the reservoir (i.e. neurons). Here we present numerical results on an optical fiber ring cavity as shown in Fig. 1 (left), which is a suitable platform to study spatially parallel reservoir computing with 1 spatial dimension.