Fully Non-Linear Neuromorphic Computing with Linear Wave Scattering

Understand

The increasing complexity of neural networks and the energy consumption associated with training and inference create a need for alternative neuromorphic approaches, e.g.

• using optics.
• Current proposals and implementations rely on physical non-linearities or opto-electronic conversion to realise the required non-linear activation function.
• However, there are significant challenges with these approaches related to power levels, control, energy-efficiency, and delays.