Artificial neural networks enabled by nanophotonics

Integrated photonic neural network based on silicon metalines

An integrated photonic neural network is proposed based on on-chip cascaded one-dimensional (1D) metasurfaces. High-contrast transmitarray metasurfaces, termed as metalines in this paper, are defined sequentially in the silicon-on-insulator substrate with a distance much larger than the operation wavelength. Matrix-vector multiplications can be accomplished in parallel and with low energy consumption due to intrinsic parallelism and low-loss of silicon metalines. The proposed on-chip whole-passive fully-optical meta-neural-network is very compact and works at the speed of light, with very low energy consumption. Various complex functions that are performed by digital neural networks can be implemented by our proposal at the wavelength of 1.55 µm. As an example, the performance of our optical neural network is benchmarked on the prototypical machine learning task of classification of handwritten digits images from the Modified National Institute of Standards and Technology (MNIST) dataset, and an accuracy comparable to the state of the art is achieved.

Fabricating acid-sensitive controlled PAA@Ag/AgCl/CN photocatalyst with reversible photocatalytic activity transformation

Achieving the intelligent controllability of the photocatalyst to the surrounding environment is a very meaningful work. Here, the polyacrylic acid (PAA) modified Ag/AgCl-40/CN composite was constructed to achieve an intelligent response of pH value. PAA exhibits hydrophilic properties at high pH value, increasing the adsorption capacity to tetracycline (TC) molecules. The morphology of PAA from contracted state to diastolic state, releasing the Ag/AgCl-40/CN catalyst. In addition, PAA modified Ag/AgCl-40/CN can prevent the loss of AgCl. The g-C₃N₄ nanosheets (CN) as a carrier enhance the dispersibility of the AgCl particles. The LSPR effects of Ag nanoparticles produce more electrons acting on photocatalytic degradation. On the results of experiment, the degradation of TC by PAA@Ag/AgCl-40/CN shows an excellent degradation activity when the high pH value. Photoluminescence spectroscopy and photocurrent demonstrate that carrier separation efficiency of PAA@Ag/AgCl-40/CN is higher than CN and Ag/AgCl-40/CN. The detection of the main active substances •O₂^- and h⁺, revealing a reasonable mechanism for the PAA@Ag/AgCl-40/CN hybrid system. This work provides a procedure to obtain smart materials that can switch photocatalytic processes.

Machine learning approach for elucidating and predicting the role of synthesis parameters on the shape and size of TiO2 nanoparticles

In the present work a series of design rules are developed in order to tune the morphology of TiO₂ nanoparticles through hydrothermal process. Through a careful experimental design, the influence of relevant process parameters on the synthesis outcome are studied, reaching to the develop predictive models by using Machine Learning methods. The models, after the validation and training, are able to predict with high accuracy the synthesis outcome in terms of nanoparticle size, polydispersity and aspect ratio. Furthermore, they are implemented by reverse engineering approach to do the inverse process, i.e. obtain the optimal synthesis parameters given a specific product characteristic. For the first time, it is presented a synthesis method that allows continuous and precise control of NPs morphology with the possibility to tune the aspect ratio over a large range from 1.4 (perfect truncated bipyramids) to 6 (elongated nanoparticles) and the length from 20 to 140 nm.

Implementing an Insect Brain Computational Circuit Using III-V Nanowire Components in a Single Shared Waveguide Optical Network

Recent developments in photonics include efficient nanoscale optoelectronic components and novel methods for subwavelength light manipulation. Here, we explore the potential offered by such devices as a substrate for neuromorphic computing. We propose an artificial neural network in which the weighted connectivity between nodes is achieved by emitting and receiving overlapping light signals inside a shared quasi 2D waveguide. This decreases the circuit footprint by at least an order of magnitude compared to existing optical solutions. The reception, evaluation, and emission of the optical signals are performed by neuron-like nodes constructed from known, highly efficient III-V nanowire optoelectronics. This minimizes power consumption of the network. To demonstrate the concept, we build a computational model based on an anatomically correct, functioning model of the central-complex navigation circuit of the insect brain. We simulate in detail the optical and electronic parts required to reproduce the connectivity of the central part of this network using previously experimentally derived parameters. The results are used as input in the full model, and we demonstrate that the functionality is preserved. Our approach points to a general method for drastically reducing the footprint and improving power efficiency of optoelectronic neural networks, leveraging the superior speed and energy efficiency of light as a carrier of information.

Reconfigurable reflective arrayed waveguide grating using optimization algorithms

In this paper we report the experimental realization of a reconfigurable reflective arrayed waveguide grating on silicon nitride technology, using optimization algorithms borrowed from machine learning applications. A dozen of band-shape responses, as well as a spectral resolution change, are demonstrated in the optical telecom C-band, alongside a proof of operation of the same device in the O-band. In the context of programmable and reconfigurable integrated photonics, this building block supports multi-wavelength/band spectral shaping of optical signals that can serve to multiple applications.

Machine Learning for Precision Breast Cancer Diagnosis and Prediction of the Nanoparticle Cellular Internalization

In the field of theranostics, diagnostic nanoparticles are designed to collect highly patient-selective disease profiles, which is then leveraged by a set of nanotherapeutics to improve the therapeutic results. Despite their early promise, high interpatient and intratumoral heterogeneities make any rational design and analysis of these theranostics platforms extremely problematic. Recent advances in deep-learning-based tools may help bridge this gap, using pattern recognition algorithms for better diagnostic precision and therapeutic outcome. Triple-negative breast cancer (TNBC) is a conundrum because of the complex molecular diversity, making its diagnosis and therapy challenging. To address these challenges, we propose a method to predict the cellular internalization of nanoparticles (NPs) against different cancer stages using artificial intelligence. Here, we demonstrate for the first time that a combination of machine-learning (ML) algorithm and characteristic cellular uptake responses for individual cancer cell types can be successfully used to classify various cancer cell types. Utilizing this approach, we can optimize the nanomaterials to get an optimum structure-internalization response for a given particle. This methodology predicted the structure-internalization response of the evaluated nanoparticles with remarkable accuracy (Q² = 0.9). We anticipate that it can reduce the effort by minimizing the number of nanoparticles that need to be tested and could be utilized as a screening tool for designing nanotherapeutics. Following this, we have proposed a diagnostic nanomaterial-based platform used to assemble a patient-specific cancer profile with the assistance of machine learning (ML). The platform is composed of eight carbon nanoparticles (CNPs) with multifarious surface chemistries that can differentiate healthy breast cells from cancerous cells and then subclassify TNBC cells vs non-TNBC cells, within the TNBC group. The artificial neural network (ANN) algorithm has been successfully used in identifying the type of cancer cells from 36 unknown cancer samples with an overall accuracy of >98%, providing potential applications in cancer diagnostics.

Direct determination of aberration functions in microscopy by an artificial neural network

Adaptive optics relies on the fast and accurate determination of aberrations but is often hindered by wavefront sensor limitations or lengthy optimization algorithms. Deep learning by artificial neural networks has recently been shown to provide determination of aberration coefficients from various microscope metrics. Here we numerically investigate the direct determination of aberration functions in the pupil plane of a high numerical aperture microscope using an artificial neural network. We show that an aberration function can be determined from fluorescent guide stars and used to improve the Strehl ratio without the need for reconstruction from Zernike polynomial coefficients.

A Novel Characterization and Performance Measurement of Memristor Devices for Synaptic Emulators in Advanced Neuro-Computing

The advanced neuro-computing field requires new memristor devices with great potential as synaptic emulators between pre- and postsynaptic neurons. This paper presents memristor devices with TiO2 Nanoparticles (NPs)/Ag(Silver) and Titanium Dioxide (TiO2) Nanoparticles (NPs)/Au(Gold) electrodes for synaptic emulators in an advanced neurocomputing application. A comparative study between Ag(Silver)- and Au(Gold)-based memristor devices is presented where the Ag electrode provides the improved performance, as compared to the Au electrode. Device characterization is observed by the Scanning Electron Microscope (SEM) image, which displays the grown electrode, while the morphology of nanoparticles (NPs) is verified by Atomic Force Microscopy (AFM). The resistive switching (RS) phenomena observed in Ag/TiO2 and Au/TiO2 shows the sweeping mechanism for low resistance and high resistance states. The resistive switching time of Au/TiO2 NPs and Ag/TiO2 NPs is calculated, while the theoretical validation of the memory window demonstrates memristor behavior as a synaptic emulator. Measurement of the capacitor-voltage curve shows that the memristor with Ag contact is a good candidate for charge storage as compared to Au. The classification of 3 × 3 pixel black/white image is demonstrated by the 3 × 3 cross bar memristor with pre- and post-neuron system. The proposed memristor devices with the Ag electrode demonstrate the adequate performance compared to the Au electrode, and may present noteworthy advantages in the field of neuromorphic computing.

Synthesized Z-scheme photocatalyst ZnO/g-C3N4 for enhanced photocatalytic reduction of CO2

ZnO/g-C₃N₄ was prepared by carrying out a simple one-step calcination process.

A Bidirectional Deep Neural Network for Accurate Silicon Color Design

Silicon nanostructure color has achieved unprecedented high printing resolution and larger color gamut than sRGB. The exact color is determined by localized magnetic and electric dipole resonance of nanostructures, which are sensitive to their geometric changes. Usually, the design of specific colors and iterative optimization of geometric parameters are computationally costly, and obtaining millions of different structural colors is challenging. Here, a deep neural network is trained, which can accurately predict the color generated by random silicon nanostructures in the forward modeling process and solve the nonuniqueness problem in the inverse design process that can accurately output the device geometries for at least one million different colors. The key results suggest deep learning is a powerful tool to minimize the computation cost and maximize the design efficiency for nanophotonics, which can guide silicon color manufacturing with high accuracy.

Laser-Inscribed Stress-Induced Birefringence of Sapphire

Birefringence of 3 × 10 - 3 is demonstrated inside cross-sectional regions of 100 μ m, inscribed by axially stretched Bessel-beam-like fs-laser pulses along the c-axis inside sapphire. A high birefringence and retardance of λ / 4 at mid-visible spectral range (green) can be achieved using stretched beams with axial extension of 30-40 μ m. Chosen conditions of laser-writing ensure that there are no formations of self-organized nano-gratings. This method can be adopted for creation of polarization optical elements and fabrication of spatially varying birefringent patterns for optical vortex generation.