Showing 1–10 of 10 results for author: George, J K

Nonlinear Optical Joint Transform Correlator for Low Latency Convolution Operations

Authors: Jonathan K. George, Maria Solyanik-Gorgone, Hangbo Yang, Chee Wei Wong, Volker J. Sorger

Abstract: Convolutions are one of the most relevant operations in artificial intelligence (AI) systems. High computational complexity scaling poses significant challenges, especially in fast-responding network-edge AI applications. Fortunately, the convolution theorem can be executed on-the-fly in the optical domain via a joint transform correlator (JTC) offering to fundamentally reduce the computational co… ▽ More Convolutions are one of the most relevant operations in artificial intelligence (AI) systems. High computational complexity scaling poses significant challenges, especially in fast-responding network-edge AI applications. Fortunately, the convolution theorem can be executed on-the-fly in the optical domain via a joint transform correlator (JTC) offering to fundamentally reduce the computational complexity. Nonetheless, the iterative two-step process of a classical JTC renders them unpractical. Here we introduce a novel implementation of an optical convolution-processor capable of near-zero latency by utilizing all-optical nonlinearity inside a JTC, thus minimizing electronic signal or conversion delay. Fundamentally we show how this nonlinear auto-correlator enables reducing the high $O(n^4)$ scaling complexity of processing two-dimensional data to only $O(n^2)$. Moreover, this optical JTC processes millions of channels in time-parallel, ideal for large-matrix machine learning tasks. Exemplary utilizing the nonlinear process of four-wave mixing, we show light processing performing a full convolution that is temporally limited only by geometric features of the lens and the nonlinear material's response time. We further discuss that the all-optical nonlinearity exhibits gain in excess of $>10^{3}$ when enhanced by slow-light effects such as epsilon-near-zero. Such novel implementation for a machine learning accelerator featuring low-latency and non-iterative massive data parallelism enabled by fundamental reduced complexity scaling bears significant promise for network-edge, and cloud AI systems. △ Less

Submitted 13 June, 2022; v1 submitted 13 February, 2022; originally announced February 2022.

An ITO Graphene hybrid integrated absorption modulator on Si-photonics for neuromorphic nonlinear activation

Authors: Rubab Amin, Jonathan K. George, Hao Wang, Rishi Maiti, Zhizhen Ma, Hamed Dalir, Jacob B. Khurgin, Volker J. Sorger

Abstract: The high demand for machine intelligence of doubling every three months is driving novel hardware solutions beyond charging of electrical wires given a resurrection to application specific integrated circuit (ASIC)-based accelerators. These innovations include photonic ASICs (P-ASIC) due to prospects of performing optical linear (and also nonlinear) operations, such as multiply-accumulate for vect… ▽ More The high demand for machine intelligence of doubling every three months is driving novel hardware solutions beyond charging of electrical wires given a resurrection to application specific integrated circuit (ASIC)-based accelerators. These innovations include photonic ASICs (P-ASIC) due to prospects of performing optical linear (and also nonlinear) operations, such as multiply-accumulate for vector matrix multiplications or convolutions, without iterative architectures. Such photonic linear algebra enables picosecond delay when photonic integrated circuits are utilized, via on-the-fly mathematics. However, the neurons full function includes providing a nonlinear activation function, knowns as thresholding, to enable decision making on inferred data. Many P-ASIC solutions performing this nonlinearity in the electronic domain, which brings challenges in terms of data throughput and delay, thus breaking the optical link and introducing increased system complexity via domain crossings. This work follows the notion of utilizing enhanced light-matter-interactions to provide efficient, compact, and engineerable electro-optic neuron nonlinearity. Here, we introduce and demonstrate a novel electro-optic device to engineer the shape of this optical nonlinearity to resemble a rectifying linear unit (ReLU) - the most-commonly used nonlinear activation function in neural networks. We combine the counter-directional transfer functions from heterostructures made out of two electro-optic materials to design a diode-like nonlinear response of the device. Integrating this nonlinearity into a photonic neural network, we show how the electrostatics of this thresholders gating junction improves machine learning inference accuracy and the energy efficiency of the neural network. △ Less

Submitted 2 September, 2021; originally announced September 2021.

All-Chalcogenide Programmable All-Optical Deep Neural Networks

Authors: Ting Yu, Xiaoxuan Ma, Ernest Pastor, Jonathan K. George, Simon Wall, Mario Miscuglio, Robert E. Simpson, Volker J. Sorger

Abstract: Deeplearning algorithms are revolutionising many aspects of modern life. Typically, they are implemented in CMOS-based hardware with severely limited memory access times and inefficient data-routing. All-optical neural networks without any electro-optic conversions could alleviate these shortcomings. However, an all-optical nonlinear activation function, which is a vital building block for optical… ▽ More Deeplearning algorithms are revolutionising many aspects of modern life. Typically, they are implemented in CMOS-based hardware with severely limited memory access times and inefficient data-routing. All-optical neural networks without any electro-optic conversions could alleviate these shortcomings. However, an all-optical nonlinear activation function, which is a vital building block for optical neural networks, needs to be developed efficiently on-chip. Here, we introduce and demonstrate both optical synapse weighting and all-optical nonlinear thresholding using two different effects in a chalcogenide material photonic platform. We show how the structural phase transitions in a wide-bandgap phase-change material enables storing the neural network weights via non-volatile photonic memory, whilst resonant bond destabilisation is used as a nonlinear activation threshold without changing the material. These two different transitions within chalcogenides enable programmable neural networks with near-zero static power consumption once trained, in addition to picosecond delays performing inference tasks not limited by wire charging that limit electrical circuits; for instance, we show that nanosecond-order weight programming and near-instantaneous weight updates enable accurate inference tasks within 20 picoseconds in a 3-layer all-optical neural network. Optical neural networks that bypass electro-optic conversion altogether hold promise for network-edge machine learning applications where decision-making in real-time are critical, such as for autonomous vehicles or navigation systems such as signal pre-processing of LIDAR systems. △ Less

Submitted 27 February, 2021; v1 submitted 20 February, 2021; originally announced February 2021.

Electronic Bottleneck Suppression in Next-generation Networks with Integrated Photonic Digital-to-analog Converters

Authors: Jiawei Meng, Mario Miscuglio, Jonathan K. George, Aydin Babakhani, Volker J. Sorger

Abstract: Digital-to-analog converters (DAC) are indispensable functional units in signal processing instrumentation and wide-band telecommunication links for both civil and military applications. Since photonic systems are capable of high data throughput and low latency, an increasingly found system limitation stems from the required domain-crossing such as digital-to-analog, and electronic-to-optical. A p… ▽ More Digital-to-analog converters (DAC) are indispensable functional units in signal processing instrumentation and wide-band telecommunication links for both civil and military applications. Since photonic systems are capable of high data throughput and low latency, an increasingly found system limitation stems from the required domain-crossing such as digital-to-analog, and electronic-to-optical. A photonic DAC implementation, in contrast, enables a seamless signal conversion with respect to both energy efficiency and short signal delay, often require bulky discrete optical components and electric-optic transformation hence introducing inefficiencies. Here, we introduce a novel coherent parallel photonic DAC concept along with an experimental demonstration capable of performing this digital-to-analog conversion without optic-electric-optic domain crossing. This design hence guarantees a linear intensity weighting among bits operating at high sampling rates, yet at a reduced footprint and power consumption compared to other photonic alternatives. Importantly, this photonic DAC could create seamless interfaces of next-generation data processing hardware for data-centers, task-specific compute accelerators such as neuromorphic engines, and network edge processing applications. △ Less

Submitted 22 December, 2019; v1 submitted 3 November, 2019; originally announced November 2019.

Journal ref: Advanced Photonics Research 2020, 2000033

A Lateral MOS-Capacitor Enabled ITO Mach-Zehnder Modulator for Beam Steering

Authors: Rubab Amin, Rishi Maiti, Jonathan K. George, Xiaoxuan Ma, Zhizhen Ma, Hamed Dalir, Mario Miscuglio, Volker J. Sorger

Abstract: Here, we experimentally demonstrate an Indium Tin Oxide (ITO) Mach-Zehnder interferometer heterogeneously integrated in silicon photonics. The phase shifter section is realized in a novel lateral MOS configuration, which, due to favorable electrostatic overlap, leads to efficient modulation (VπL = 63 Vum). This is achieved by (i) selecting a strong index changing material (ITO) and (ii) improving… ▽ More Here, we experimentally demonstrate an Indium Tin Oxide (ITO) Mach-Zehnder interferometer heterogeneously integrated in silicon photonics. The phase shifter section is realized in a novel lateral MOS configuration, which, due to favorable electrostatic overlap, leads to efficient modulation (VπL = 63 Vum). This is achieved by (i) selecting a strong index changing material (ITO) and (ii) improving the field overlap as verified by the electrostatic field lines. Furthermore, we show that this platform serves as a building block in an endfire silicon photonics optical phased array (OPA) with a half-wavelength pitch within the waveguides with anticipated performance, including narrow main beam lobe (<3°) and >10 dB suppression of the side lobes, while electrostatically steering the emission profile up to plus/minus 80°, and if further engineered, can lead not only towards nanosecond-fast beam steering capabilities in LiDAR systems but also in holographic display, free-space optical communications, and optical switches. △ Less

Submitted 1 November, 2019; v1 submitted 1 June, 2019; originally announced July 2019.

All-optical Nonlinear Activation Function for Photonic Neural Networks

Authors: Mario Miscuglio, Armin Mehrabian, Zibo Hu, Shaimaa I. Azzam, Jonathan K. George, Alexander V. Kildishev, Matthew Pelton, Volker J. Sorger

Abstract: With the recent successes of neural networks (NN) to perform machine-learning tasks, photonic-based NN designs may enable high throughput and low power neuromorphic compute paradigms since they bypass the parasitic charging of capacitive wires. Thus, engineering data-information processors capable of executing NN algorithms with high efficiency is of major importance for applications ranging from… ▽ More With the recent successes of neural networks (NN) to perform machine-learning tasks, photonic-based NN designs may enable high throughput and low power neuromorphic compute paradigms since they bypass the parasitic charging of capacitive wires. Thus, engineering data-information processors capable of executing NN algorithms with high efficiency is of major importance for applications ranging from pattern recognition to classification. Our hypothesis is therefore, that if the time-limiting electro-optic conversion of current photonic NN designs could be postponed until the very end of the network, then the execution time of the photonic algorithm is simple the delay of the time-of-flight of photons through the NN, which is on the order of picoseconds for integrated photonics. Exploring such all-optical NN, in this work we discuss two independent approaches of implementing the optical perceptrons nonlinear activation function based on nanophotonic structures exhibiting i) induced transparency and ii) reverse saturated absorption. Our results show that the all-optical nonlinearity provides about 3 and 7 dB extinction ratio for the two systems considered, respectively, and classification accuracies of an exemplary MNIST task of 97% and near 100% are found, which rivals that of software based trained NNs, yet with ignored noise in the network. Together with a developed concept for an all-optical perceptron, these findings point to the possibility of realizing pure photonic NNs with potentially unmatched throughput and even energy consumption for next generation information processing hardware. △ Less

Submitted 6 October, 2018; v1 submitted 2 October, 2018; originally announced October 2018.

Integrated All-Optical Fast Fourier Transform: Design and Sensitivity Analysis

Authors: Hani Nejadriahi, David HillerKuss, Jonathan K. George, Volker J. Sorger

Abstract: The fast Fourier transform, FFT, is a useful and prevalent algorithm in signal processing. It characterizes the spectral components of a signal, or is used in combination with other operations to perform more complex computations such as filtering, convolution, and correlation. Digital FFTs are limited in speed by the necessity of moving charge within logic gates. An analog temporal FFT in fiber o… ▽ More The fast Fourier transform, FFT, is a useful and prevalent algorithm in signal processing. It characterizes the spectral components of a signal, or is used in combination with other operations to perform more complex computations such as filtering, convolution, and correlation. Digital FFTs are limited in speed by the necessity of moving charge within logic gates. An analog temporal FFT in fiber optics has been demonstrated with highest data bandwidth. However, the implementation with discrete fiber optic FFT components is bulky. Here, we present and analyze a design of an optical FFT in Silicon photonics and evaluate its performance with respect to variations in phase and amplitude. We discuss the impact of the deployed devices on the FFTs transfer function quality as defined by the transmission output power as a function of frequency, detuning phase, optical delay, and loss. △ Less

Submitted 31 October, 2017; originally announced November 2017.

Identifying Mirror Symmetry Density with Delay in Spiking Neural Networks

Authors: Jonathan K. George, Cesare Soci, Volker J. Sorger

Abstract: The ability to rapidly identify symmetry and anti-symmetry is an essential attribute of intelligence. Symmetry perception is a central process in human vision and may be key to human 3D visualization. While previous work in understanding neuron symmetry perception has concentrated on the neuron as an integrator, here we show how the coincidence detecting property of the spiking neuron can be used… ▽ More The ability to rapidly identify symmetry and anti-symmetry is an essential attribute of intelligence. Symmetry perception is a central process in human vision and may be key to human 3D visualization. While previous work in understanding neuron symmetry perception has concentrated on the neuron as an integrator, here we show how the coincidence detecting property of the spiking neuron can be used to reveal symmetry density in spatial data. We develop a method for synchronizing symmetry-identifying spiking artificial neural networks to enable layering and feedback in the network. We show a method for building a network capable of identifying symmetry density between sets of data and present a digital logic implementation demonstrating an 8x8 leaky-integrate-and-fire symmetry detector in a field programmable gate array. Our results show that the efficiencies of spiking neural networks can be harnessed to rapidly identify symmetry in spatial data with applications in image processing, 3D computer vision, and robotics. △ Less

Submitted 25 August, 2017; originally announced September 2017.

Comments: 8 pages, 8 figures

Scattering and Absorption Control in Biocompatible Fibers towards Equalized Photobiomodulation

Authors: Jonathan K. George, Hamed Haghshenas, Duncan DHemecourt, Wei Zhu, Lijie Zhang, Volker J. Sorger

Abstract: Transparent tissue scaffolds enable illumination of growing tissue to accelerate cell proliferation and improve other cell functions through photobiomodulation. The biphasic dose response of cells exposed to photobiomodulating light dictates that the illumination to be evenly distributed across the scaffold such that the cells are neither under nor over exposed to light. However, equalized illumin… ▽ More Transparent tissue scaffolds enable illumination of growing tissue to accelerate cell proliferation and improve other cell functions through photobiomodulation. The biphasic dose response of cells exposed to photobiomodulating light dictates that the illumination to be evenly distributed across the scaffold such that the cells are neither under nor over exposed to light. However, equalized illumination has not been sufficiently addressed. Here we analyze and experimentally demonstrate spatially equalizing illumination by three methods, namely, engineered surface scattering, reflection by a gold mirror, and traveling-waves in a ring mesh. Our results show that nearly equalized illumination is achievable by controlling the light scattering-to-loss ratio. This demonstration opens opportunities for dose-optimized photobiomodulation in tissue regeneration. △ Less

Submitted 7 December, 2016; originally announced December 2016.

Comments: 10 pages, 4 figures

Graphene-based Solitons for Spatial Division Multiplexed Switching

Authors: Jonathan K. George, Volker J. Sorger

Abstract: Spatial division multiplexing utilizes the directionality of lights propagating k-vector to separate it into distinct spatial directions. Here we show that the anisotropy of orthogonal spatial Solitons propagating in a single Graphene monolayer results in phase-based multiplexing. We use the self- confinement properties of spatial Solitons to increase the usable density of states (DOS) of this swi… ▽ More Spatial division multiplexing utilizes the directionality of lights propagating k-vector to separate it into distinct spatial directions. Here we show that the anisotropy of orthogonal spatial Solitons propagating in a single Graphene monolayer results in phase-based multiplexing. We use the self- confinement properties of spatial Solitons to increase the usable density of states (DOS) of this switching-system. Furthermore, we show that crossing two orthogonal Solitons exhibit a low (0.035 dB) mutual disturbance from another enabling independent k-vector switching. The efficient utilization of the DOS and multiplexing in real-space enables data processing parallelism with applications in optical networking and computing. △ Less

Submitted 7 December, 2016; originally announced December 2016.

Comments: 8 pages, 4 figures