Nonlinear optical encoding enabled by recurrent linear scattering

Abstract

Optical information processing and computing can potentially offer enhanced performance, scalability and energy efficiency. However, achieving nonlinearity-a critical component of computation-remains challenging in the optical domain. Here we introduce a design that leverages a multiple-scattering cavity to passively induce optical nonlinear random mapping with a continuous-wave laser at a low power. Each scattering event effectively mixes information from different areas of a spatial light modulator, resulting in a highly nonlinear mapping between the input data and output pattern. We demonstrate that our design retains vital information even when the readout dimensionality is reduced, thereby enabling optical data compression. This capability allows our optical platforms to offer efficient optical information processing solutions across applications. We demonstrate our design's efficacy across tasks, including classification, image reconstruction, keypoint detection and object detection, all of which are achieved through optical data compression combined with a digital decoder. In particular, high performance at extreme compression ratios is observed in real-time pedestrian detection. Our findings open pathways for novel algorithms and unconventional architectural designs for optical computing.

14 pages, 4 figures

References in corpus (11)

• Photonics for artificial intelligence and neuromorphic computing
• 11 TeraFLOPs per second photonic convolutional accelerator for deep learning optical neural networks
• Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems
• Light fields in complex media: mesoscopic scattering meets wave control
• Image sensing with multilayer, nonlinear optical neural networks
• Spectrally-Encoded Single-Pixel Machine Vision Using Diffractive Networks
• Deep Learning with Coherent VCSEL Neural Networks
• Scalable Optical Learning Operator
• Free-space optical neural network based on thermal atomic nonlinearity
• Small-brain neural networks rapidly solve inverse problems with vortex Fourier encoders
• Adversarial Robustness by Design through Analog Computing and Synthetic Gradients

Cited by in corpus (9)

• Optical Generative Models
• Programmable metasurfaces for future photonic artificial intelligence
• Transferable polychromatic optical encoder for neural networks
• Principles and Metrics of Extreme Learning Machines Using a Highly Nonlinear Fiber
• Nonlinear Inference Capacity of Fiber-Optical Extreme Learning Machines
• Massively parallel and universal approximation of nonlinear functions using diffractive processors
• Demonstration of a Tunable Non-Hermitian Nonlinear Microwave Dimer
• A universal approximation theorem for nonlinear resistive networks
• Shannon Entropy Helps Optimize the Performance of a Frequency-Multiplexed Extreme Learning Machine