We build and use a neuromorphic wave computer established on soliton-based spectral broadening. Featuring low-power consumption, robustness, and scalability, our approach enables the emulation of a variety of neural networks using a single system setting, opening the way to opportunities in sustainable photonic computing and machine learning applications.

Supercontinuum (SC) generation holds significant importance across diverse scientific and technological applications, thanks to its unique temporal-spectral properties. The authors proposed an innovative approach to supercontinuum engineering, leveraging the assistance of machine learning. We realized an on-demand engineering of SC with arbitrary bandwidth (10-70 nm) and spectral patterns of high-order solitons (N ≤ 4). This work opens new avenues for tailored and adaptive supercontinuum generation, within a highly complex nonlinear system.

Neuromorphic photonics leverages the advantages of photonics such as low latency, high bandwidth, and energy efficiency to achieve superior computing performance. A novel reconfigurable photonic signal processor addresses a range of multipurpose applications that pose significant challenges for conventional digital processors. In this talk, we will highlight recent progress in neuromorphic photonic integrated circuits and their applications in signal unscrambling, RF blind source separation and tensor- core processing.

Light’s ability to perform massive linear operations parallelly has recently inspired numerous demonstrations of optics-assisted artificial neural networks (ANN). However, a clear advantage of optics over purely digital ANN in a system-level has not yet been established. While linear operations can indeed be optically performed very efficiently, the lack of nonlinearity and signal regeneration require high-power, low-latency signal transduction between optics and electronics. Additionally, a large power is needed for lasers and photodetectors, which are often neglected in the calculation of the total energy consumption. Here, instead of mapping traditional digital operations to optics, we co-designed a hybrid optical-digital ANN, that operates on incoherent light, and is thus amenable to operations under ambient light. Keeping the latency and power constant between a purely digital ANN and a hybrid optical-digital ANN, we identified a low-power/ latency regime, where an optical encoder provides higher classification accuracy than a purely digital ANN. We estimate our optical encoder enables ~ 10kHz rate operation of a hybrid ANN with a power of only 23mW. However, in that regime, the overall classification accuracy is lower than what is achievable with higher power and latency. Our results indicate that optics can be advantageous over digital ANN in applications, where the overall performance of the ANN can be relaxed to prioritize lower power and latency.