Wave-based extreme deep learning based on non-linear time-Floquet entanglement

Abstract

Wave-based analog signal processing holds the promise of extremely fast, on-the-fly, power-efficient data processing, occurring as a wave propagates through an artificially engineered medium. Yet, due to the fundamentally weak non-linearities of traditional wave materials, such analog processors have been so far largely confined to simple linear projections such as image edge detection or matrix multiplications.Complex neuromorphic computing tasks, which inherently require strong non-linearities, have so far remained out-of-reach of wave-based solutions, with a few attempts that implemented non-linearities on the digital front, or used weak and inflexible non-linear sensors, restraining the learning performance.Here, we tackle this issue by demonstrating the relevance of Time-Floquet physics to induce a strong non-linear entanglement between signal inputs at different frequencies, enabling a power-efficient and versatile wave platform for analog extreme deep learning involving a single, uniformly modulated dielectric layer and a scattering medium. We prove the efficiency of the method for extreme learning machines and reservoir computing to solve a range of challenging learning tasks, from forecasting chaotic time series to the simultaneous classification of distinct datasets. Our results open the way for wave-based machine learning with high energy efficiency, speed and scalability. IntroductionRecently, artificial intelligence (AI) systems based on advanced machine learning algorithms have attracted a surge of interest for their potential applications in processing the information hidden in large datasets 1, 2 .Wave-based analog implementations of these schemes, exploiting microwave or optical neural networks,

Intelligent meta-imagers: From compressed to learned sensing

“…236,237 Ideas for adding non-linearities include the use of diode-loaded ports 238 and input-signal-dependent time modulation. 239 The bulkiness of 3D volumetric scattering enclosures appears at first sight a major obstacle to integrability. However, on the one hand, integrability can be accomplished by utilizing flat quasi-2D programmable chaotic cavities.…”