Reconfigurable Low-Threshold All-Optical Nonlinear Activation Functions Based on an Add-Drop Silicon Microring Resonator

Yu, Weizhen; Zheng, Shuang; Zhao, Zhenyu; Wang, Bin; Zhang, Weifeng

doi:10.1109/jphot.2022.3219246

Cited by 4 publications

(2 citation statements)

References 48 publications

(54 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Photonic activation functions can be built using all-optical or electro-optical devices. The all-optical approach benefits from a lower latency response than the optical-electrical-optical (O-E-O) conversion required by electro-optical devices [18][19][20][21]. However, current all-optical activation functions face an essential challenge since both the activation function's controlled (output) and controlling (input) optical signals affect its response.…”

Section: Introductionmentioning

confidence: 99%

On-chip electro-optical spiking VO₂/Si device with an inhibitory leaky integrate-and-fire response

Morcillo,

Kilders,

Parra

2024

Preprint

View full text Add to dashboard Cite

Integrated photonic neuromorphic computing has emerged as an appealing technology to replicate human brain functionalities by leveraging the benefits of integrated photonics. In this context, spiking neural networks (SNNs) offer a highly energy-efficient artificial neural network due to the event-based information processing. In this work, we propose an electro-optical spiking device for the silicon photonics platform based on a VO₂/Si waveguide and a meander-shaped microheater. A leaky integrate-and-fire (LIF) response with inhibitory optical spiking is achieved by harnessing the temperature-driven insulator-metal transition (IMT) of VO₂ combined with thermal dynamics. Hence, our proposal would not require complex electronic circuitry bottlenecking the neural network. Our numerical simulations demonstrate firing upon milliwatt and nanosecond range postsynaptic pulses and adjustable spike characteristics in the time domain with sub-microsecond values. Therefore, our spiking device could enable scalable and energy-efficient photonic-based SNNs.

show abstract

Section: Introductionmentioning

confidence: 99%

On-chip electro-optical spiking VO₂/Si device with an inhibitory leaky integrate-and-fire response

Morcillo,

Kilders,

Parra

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The nonlinear operation components developed thus far can be classi ed into two main categories based on their operating mechanism: optoelectronic conversion 9,10 and all-optical components 11,12 . However, optoelectronic conversion methods face challenges in terms of processing speed, complexities and energy e ciency, as they necessitate signal conversion among devices such as photodetector (PD), transimpedance ampli er (TIA) and modulator, which demand supplementary electrical energy for sustained operation 13 .…”

Section: Introductionmentioning

confidence: 99%

Ultrafast Silicon Optical Nonlinear Activator for Neuromorphic Computing

Yan

Zhou

Liu

et al. 2023

Preprint

View full text Add to dashboard Cite

Optical neural networks (ONNs) have shown great promise in overcoming the speed and efficiency bottlenecks of artificial neural networks (ANNs). However, the absence of high-speed, energy-efficient nonlinear activators significantly impedes the advancement of ONNs and their extension to ultrafast application scenarios like autonomous vehicles and real-time intelligent signal processing. In this work, we designed and fabricated a novel silicon-based ultrafast all-optical nonlinear activator, leveraging the hybrid integration of silicon slot waveguides, plasmonic slot waveguides, and monolayer graphene. We utilized double-balanced detection and synchronous pump-probe measurement techniques to experimentally evaluate the static and dynamic characteristics of the activators, respectively. Exploiting the exceptional picosecond scale photogenerated carrier relaxation time of graphene, the response time of the activator is markedly reduced to ~93.6 ps. This response time is approximately five times faster than electronic neural networks, establishing our all-optical activator as the fastest known in silicon photonics to our knowledge. Moreover, the all-optical nonlinear activator holds a low threshold power of 5.49 mW and a corresponding power consumption per activation of 0.51 pJ. Furthermore, we confirm its feasibility and capability for use in ONNs by simulation, achieving a high accuracy of 96.8% for MNIST handwritten digit recognition and a mean absolute error of less than 0.1 dB for optical signal-to-noise ratio monitoring of high-speed optical signals. This breakthrough in speed and energy efficiency of all-optical nonlinear activators opens the door to significant improvements in the performance and applicability of ONNs, ushering in a new era of advanced artificial intelligence technologies with enormous potential.

show abstract

Low-threshold all-optical nonlinear activation function based on injection locking in distributed feedback laser diodes

Liu¹,

Xu²,

Ma³

et al. 2023

Opt. Lett.

View full text Add to dashboard Cite

We experimentally demonstrate an all-optical nonlinear activation unit based on the injection-locking effect of distributed feedback laser diodes (DFB-LDs). The nonlinear carrier dynamics in the unit generates a low-threshold nonlinear activation function with optimized operating conditions. The unit can operate at a low threshold of −15.86 dBm and a high speed of 1 GHz, making it competitive among existing optical nonlinear activation approaches. We apply the unit to a neural network task of solving the second-order ordinary differential equation. The fitting error is as low as 0.0034, verifying the feasibility of our optical nonlinear activation approach. Given that the large-scale fan-out of optical neural networks (ONNs) will significantly reduce the optical power in one channel, our low-threshold scheme is suitable for the development of high-throughput ONNs.

show abstract

Reconfigurable Low-Threshold All-Optical Nonlinear Activation Functions Based on an Add-Drop Silicon Microring Resonator

Cited by 4 publications

References 48 publications

On-chip electro-optical spiking VO₂/Si device with an inhibitory leaky integrate-and-fire response

On-chip electro-optical spiking VO₂/Si device with an inhibitory leaky integrate-and-fire response

Ultrafast Silicon Optical Nonlinear Activator for Neuromorphic Computing

Low-threshold all-optical nonlinear activation function based on injection locking in distributed feedback laser diodes

Contact Info

Product

Resources

About