Handwritten Digit Classification Research Articles

Digital-analog quantum learning on Rydberg atom arrays

Abstract We propose hybrid digital-analog learning algorithms on Rydberg atom arrays, combining the potentially practical utility and near-term realizability of quantum learning with the rapidly scaling architectures of neutral atoms. Our construction requires only single-qubit operations in the digital setting and global driving according to the Rydberg Hamiltonian in the analog setting. We perform a comprehensive numerical study of our algorithm on both classical and quantum data, given respectively by handwritten digit classification and unsupervised quantum phase boundary learning. We show in the two representative problems that digital-analog learning is not only feasible in the near term, but also requires shorter circuit depths and is more robust to realistic error models as compared to digital learning schemes. Our results suggest that digital-analog learning opens a promising path towards improved variational quantum learning experiments in the near term.

Metasurface-Based Image Classification Using Diffractive Deep Neural Network

The computer-assisted inverse design of photonic computing, especially by leveraging artificial intelligence algorithms, offers great convenience to accelerate the speed of development and improve calculation accuracy. However, traditional thickness-based modulation methods are hindered by large volume and difficult fabrication process, making it hard to meet the data-driven requirements of flexible light modulation. Here, we propose a diffractive deep neural network (D2NN) framework based on a three-layer all-dielectric phased transmitarray as hidden layers, which can perform the classification of handwritten digits. By tailoring the radius of a silicon nanodisk of a meta-atom, the metasurface can realize the phase profile calculated by D2NN and maintain a relative high transmittance of 0.9 at a wavelength of 600 nm. The designed image classifier consists of three layers of phase-only metasurfaces, each of which contains 1024 units, mimicking a fully connected neural network through the diffraction of light fields. The classification task of handwriting digits from the ‘0’ to ‘5’ dataset is verified, with an accuracy of over 90% on the blind test dataset, as well as demonstrated by the full-wave simulation. Furthermore, the performance of the more complex animal image classification task is also validated by increasing the number of neurons to enhance the connectivity of the neural network. This study may provide a possible solution for practical applications such as biomedical detection, image processing, and machine vision based on all-optical computing.

Scalable parallel photonic processing unit for various neural network accelerations

In recent years, integrated optical processing units (IOPUs) have demonstrated advantages in energy efficiency and computational speed for neural network inference applications. However, limited by optical integration technology, the practicality and versatility of IOPU face serious challenges. In this work, a scalable parallel photonic processing unit (SPPU) for various neural network accelerations based on high-speed phase modulation is proposed and implemented on a silicon-on-insulator platform, which supports parallel processing and can switch between multiple computational paradigms simply and without latency to infer different neural network structures, enabling to maximize the utility of on-chip components. The SPPU adopts a scalable and process-friendly architecture design, with a preeminent photonic-core energy efficiency of 0.83 TOPS/W, two to ten times higher than existing integrated solutions. In the proof-of-concept experiment, a convolutional neural network (CNN), a residual CNN, and a recurrent neural network (RNN) are all implemented on our photonic processor to handle multiple tasks of handwritten digit classification, signal modulation format recognition, and review emotion recognition. The SPPU achieves multi-task parallel processing capability, serving as a promising and attractive research route to maximize the utility of on-chip components under the constraints of integrated technology, which helps to make IOPU more practical and universal.

Comparative analysis of CNN models for handwritten digit recognition

The paper discusses the subject of convolutional neural networks used for handwritten digit classification. The purpose of the research is to evaluate the accuracy, performance, training, and classification time of three OCR networks (VGG-16, VGG-19 and AlexNet) and compare them with each other while selecting the most optimal one. The popular MNIST dataset of 70,000 images was used for the study. For each model, a preliminary study was conducted to determine the optimal parameters in the form of the number of input data and number of training epochs. The result of the work indicates that, despite the longer training and classification time, the AlexNet model achieved the highest precision, recall, and F1-score, indicating its ability to effectively classify images.

On-chip photoelectric hybrid convolutional accelerator based on Bragg grating array

We propose an on-chip photoelectric hybrid convolution accelerator based on Bragg grating array. The weight of the convolution kernel can be adjusted by controlling the central wavelengths of the Bragg grating array. We conducted simulation verification of the functionality and scalability of this on-chip photoelectric hybrid convolution accelerator. Subsequently, we constructed a neural network model to conduct handwritten digit classification simulations using this accelerator, achieving a simulation accuracy of 93.99%. Finally, the concept of the proposed on-chip photoelectric hybrid convolution accelerator is successfully verified. Due to the merits of Bragg grating, the proposed scheme paves the way for realizing high-performance on-chip optical neural networks.

Terahertz spoof plasmonic neural network for diffractive information recognition and processing

All-optical diffractive neural networks, as analog artificial intelligence accelerators, leverage parallelism and analog computation for complex data processing. However, their low space transmission efficiency or large spatial dimensions hinder miniaturization and broader application. Here, we propose a terahertz spoof plasmonic neural network on a planar diffractive platform for direct multi-target recognition. Our approach employs a spoof surface plasmon polariton coupler array to construct a diffractive network layer, resulting in a compact, efficient, and easily integrable architecture. We designed three schemes: basis vector classification, multi-user recognition, and MNIST handwritten digit classification. Experimental results reveal that the terahertz spoof plasmonic neural network successfully classifies basis vectors, recognizes multi-user orientation information, and directly processes handwritten digits using a designed input framework comprising a metal grating array, transmitters, and receivers. This work broadens the application of terahertz plasmonic metamaterials, paving the way for terahertz on-chip integration, intelligent communication, and advanced computing systems.

Ultra-fast GST-based optical neuron for the implementation of integrated photonic neural networks

A fast-integrated optical neuron is proposed based on a micro-ring resonator structure. The GST phase change material is used in the MRR structure, and neuron switching is achieved by the electrical actuation of this material. The nonlinear activation function of the neuron is extracted through three-dimensional electrical and optical simulations. The neuron is then employed in a three-layer perceptron network using the extracted activation function, and its performance is examined in the classification of handwritten digits. The overall performance of the proposed neuron is compared with the previous optical neural networks. Simulations show an ultra-fast switching time of 200fs, which is one order of magnitude faster than the best-reported value, and an accuracy of 98.9% (in the classification of MNIST handwritten digits dataset) which is among the best results, while the estimated neuron footprint is relatively small.

Ultracompact and multifunctional integrated photonic platform.

Realizing a multifunctional integrated photonic platform is one of the goals for future optical information processing, which usually requires large size to realize due to multiple integration challenges. Here, we realize a multifunctional integrated photonic platform with ultracompact footprint based on inverse design. The photonic platform is compact with 86 inverse designed-fixed couplers and 91 phase shifters. The footprint of each coupler is 4 μm by 2 μm, while the whole photonic platform is 3 mm by 0.2 mm-one order of magnitude smaller than previous designs. One-dimensional Floquet Su-Schrieffer-Heeger model and Aubry-André-Harper model are performed with measured fidelities of 97.90 (±0.52) % and 99.34 (±0.44) %, respectively. We also demonstrate a handwritten digits classification task with the test accuracy of 87% using on-chip training. Moreover, the scalability of this platform has been proved by demonstrating more complex computing tasks. This work provides an effective method to realize an ultrasmall integrated photonic platform.

Cultural Algorithm Guided Policy Gradient with Parameter Exploration

This study explores the integration of cultural algorithms (CA) with the Policy Gradients with Parameter-Based Exploration (PGPE) algorithm for the task of MNIST hand-written digit classification within the EvoJAX framework. The PGPE algorithm is enhanced by incorporating a belief space, consisting on Domain, Situational, and History knowledge sources (KS), to guide the search process and improve convergence speed. The PGPE algorithm, implemented within the EvoJAX framework, can efficiently find an optimal parameter-space policy for the MNIST task. However, increasing the complexity of the task and policy space, such as the CheXpert dataset and DenseNet, requires a more sophisticated approach to efficiently navigate the search space. We introduce CA-PGPE, a novel approach that integrates CA with PGPE to guide the search process and improve convergence speed. Future work will focus on incorporating exploratory knowledge sources and evaluate the enhanced CA-PGPE algorithm on more complex datasets and model architectures, such as CIFAR-10 and CheXpert with DenseNet.

CNN Module on MNIST Dataset for Written Digit Classification

Abstract: Convolutional Neural Networks (CNNs) have become indispensable tools in the realm of image classification, particularly in tasks like handwritten digit recognition. In this comprehensive study, we delve into the intricate world of CNN modules as applied to the MNIST dataset, a cornerstone benchmark in machine learning. Our research aims to meticulously assess the performance of diverse CNN architectures, encompassing variations in depth, convolutional layer configurations, pooling strategies, and regularization techniques. Through exhaustive experimentation and meticulous analysis, we endeavor to offer profound insights into the nuanced strengths and limitations of different CNN modules for the task of handwritten digit classification on the MNIST dataset. By elucidating the intricacies of CNN architecture, we endeavor to contribute to the advancement of image classification methodologies, particularly in domains where labeled data is scarce and precision is paramount.

Comparison of Bagging and Sparcity Methods for Connectivity Reduction in Spiking Neural Networks with Memristive Plasticity

Developing a spiking neural network architecture that could prospectively be trained on energy-efficient neuromorphic hardware to solve various data analysis tasks requires satisfying the limitations of prospective analog or digital hardware, i.e., local learning and limited numbers of connections, respectively. In this work, we compare two methods of connectivity reduction that are applicable to spiking networks with local plasticity; instead of a large fully-connected network (which is used as the baseline for comparison), we employ either an ensemble of independent small networks or a network with probabilistic sparse connectivity. We evaluate both of these methods with a three-layer spiking neural network, which are applied to handwritten and spoken digit classification tasks using two memristive plasticity models and the classical spike time-dependent plasticity (STDP) rule. Both methods achieve an F1-score of 0.93–0.95 on the handwritten digits recognition task and 0.85–0.93 on the spoken digits recognition task. Applying a combination of both methods made it possible to obtain highly accurate models while reducing the number of connections by more than three times compared to the basic model.

The Image Classification Algorithm Was Implemented on The MNIST Data Set

In the current era of rapid development of science and technology, the recognition and classification of digital images is the key to solving many problems, such as the application of license plate recognition, document digitization, and remote sensing image surface classification. Based on the MNIST handwritten numerical data set collected by the National Institute of Standards and Technology (NIST), this report uses Python language and PyTorch programming framework to construct a convolutional neural network (CNN) structure and practice and experience the image classification of handwritten digits in the MNIST data set.

Enhancing plasticity in optoelectronic artificial synapses: A pathway to efficient neuromorphic computing

The continuous growth in artificial intelligence and high-performance computing has necessitated the development of efficient optoelectronic artificial synapses crucial for neuromorphic computing (NC). Ga2O3 is an emerging wide-bandgap semiconductor with high deep ultraviolet absorption, tunable persistent photoconductivity, and excellent stability toward electric fields, making it a promising component for optoelectronic artificial synapses. Currently reported Ga2O3 optoelectronic artificial synapses often suffer from complex fabrication processes and potential room for improvement due to plasticity. To address the issue of low device plasticity and practical application scenarios, we present an amorphous Ga2O3 (α-GaOx) flexible optoelectronic artificial synapse. This synapse modulates light stimulus signals using electron/oxygen vacancies and optical stimulation and operates as a visual storage device for information processing. We investigate the improvement of the optoelectronic synapses' plasticity by controlling the number of oxygen vacancies via a plasma treatment method and demonstrate its effective application in a three-layer backpropagation neural network for handwritten digit classification. Under the same stimulus conditions, the synaptic weight of samples treated with Ar plasma exhibits a higher rate of change, with the current levels increasing by 2–3 orders of magnitude, achieving greater plasticity. The improved optoelectronic synapses achieved an accuracy of 93.34%/94%, demonstrating their potential as efficient computing solutions and insights for future applications in NC chips.

(Invited) Nanoscale Devices for Accelerating AI Algorithms

The energy consumption of training and deploying state-of-the-art artificial intelligence (AI) models has experienced exponential growth, driven by increasing model parameters and the necessary training data. Recent models demand over 1 billion PetaFLOPs of total computation for training, potentially taking weeks to complete. With current GPU performance at approximately 1-3 TOPS/W, GPU training energy consumption alone can exceed 100,000 kWh, equivalent to the monthly energy expenditure of 100 US households. Over the next decade, these models are expected to scale up further, driving total computing energy to constitute a significant portion of global consumption. In this talk, we describe both synaptic and neuronal devices that can accelerate AI algorithms with potentially multiple orders of magnitude improvement in power efficiency.First, we describe an oscillatory retinal neuron (ORN) that directly converts incident DC light into voltage spikes. Coupled arrays of this device result in an imager that carries out in-sensor processing while capturing an image. Uniquely, the conversion from input light to output voltage spikes occurs without external power. When coupled in arrays, the neighboring neurons interact with each other to influence the spiking frequency spectrum. This allows the arrays to carry out frequency multiplexed computation on an input image. It is shown that this approach can achieve>20,000 TOPS/W, multiple orders of magnitude greater than the current approaches. Theory and simulation is used to elucidate how coupled ORNs carry out computation on an input image. When coupled, each output frequency band encodes a unique computation on the input image. By tuning the coupling impedances and the frequency bands, user defined computations can be carried out on the input. We experimentally show that this can be carried out with a 3x3 array that demonstrates simultaneous edge detection, intensity filtering, image segmentation and other functions. This hardware is then used to demonstrate improvement in MNIST handwritten digit classification accuracy over a traditional imager connected to a fully connected network.Next, we describe spiking synaptic devices that can be directly fabricated in the back-end of line of CMOS devices. These devices consist of an InP transistor channel with an engineered gate stack. Using a uniform gate insulator, we can demonstrate behaviors of biological synapses, such as potentiation, depression, spike number dependent plasticity, and spike timing dependent plasticity. By introducing a heterostructured gate insulator, it is shown that short-term to long-term memory transitions can be designed into the device. Finally, by using a transparent gate, an in-sensor synaptic phototransistor is demonstrated and the performance of these devices at a system level is demonstrated.

QTN-VQC: an end-to-end learning framework for quantum neural networks

This work focuses on investigating an end-to-end learning approach for quantum neural networks (QNN) on noisy intermediate-scale quantum devices. The proposed model combines a quantum tensor network (QTN) with a variational quantum circuit (VQC), resulting in a QTN-VQC architecture. This architecture integrates a QTN with a horizontal or vertical structure related to the implementation of quantum circuits for a tensor-train network. The study provides theoretical insights into the quantum advantages of the end-to-end learning pipeline based on QTN-VQC from two perspectives. The first perspective refers to the theoretical understanding of QTN-VQC with upper bounds on the empirical error, examining its learnability and generalization powers; The second perspective focuses on using the QTN-VQC architecture to alleviate the Barren Plateau problem in the training stage. Our experimental simulation on CPU/GPUs is performed on a handwritten digit classification dataset to corroborate our proposed methods in this work.

Harnessing 2D Ruddlesden-Popper Perovskite with Polar Organic Cation for Ultrasensitive Multibit Nonvolatile Transistor-Type Photomemristors.

Photomemristors have been regarded as one of the most promising candidates for next-generation hardware-based neuromorphic computing due to their potentials of fast data transmission and low power consumption. However, intriguingly, so far, photomemristors seldom display truly nonvolatile memory characteristics with high light sensitivity. Herein, we demonstrate ultrasensitive photomemristors utilizing two-dimensional (2D) Ruddlesden-Popper (RP) perovskites with a highly polar donor-acceptor-type push-pull organic cation, 4-(5-(2-aminoethyl)thiophen-2-yl)benzonitrile+ (EATPCN+), as charge-trapping layers. High linearity and almost zero-decay retention are observed in (EATPCN)2PbI4 devices, which are very distinct from that of the traditional 2D RP perovskite devices consisting of nonpolar organic cations, such as phenethylamine+ (PEA+) and octylamine+ (OA+), and traditional 3D perovskite devices consisting of methylamine+ (MA+). The 2-fold advantages, including desirable spatial crystal arrangement and engineered energetic band alignment, clarify the mechanism of superior performance in (EATPCN)2PbI4 devices. The optimized (EATPCN)2PbI4 photomemristor also shows a memory window of 87.9 V and an on/off ratio of 106 with a retention time of at least 2.4 × 105 s and remains unchanged after >105 writing-reading-erasing-reading endurance cycles. Very low energy consumptions of 1.12 and 6 fJ for both light stimulation and the reading process of each status update are also demonstrated. The extremely low power consumption and high photoresponsivity were simultaneously achieved. The high photosensitivity surpasses that of a state-of-the-art commercial pulse energy meter by several orders of magnitude. With their outstanding linearity and retention, rabbit images have been rebuilt by (EATPCN)2PbI4 photomemristors, which truthfully render the image without fading over time. Finally, by utilizing the powerful ∼8 bits of nonvolatile potentiation and depression levels of (EATPCN)2PbI4 photomemristors, the accuracies of the recognition tasks of CIFAR-10 image classification and MNIST handwritten digit classification have reached 89% and 94.8%, respectively. This study represents the first report of utilizing a functional donor-acceptor type of organic cation in 2D RP perovskites for high-performance photomemristors with characteristics that are not found in current halide perovskites.

Neuromorphic convolution with a spiking DFB-SA laser neuron based on rate coding.

We propose a neuromorphic convolution system using a photonic integrated distributed feedback laser with a saturable absorber (DFB-SA) as a photonic spiking neuron. The experiments reveal that the DFB-SA laser can encode different stimulus intensities at different frequencies, similar to biological neurons. Based on this property, optical inputs are encoded into rectangular pulses of varying intensities and injected into the DFB-SA laser, enabling the convolution results to be represented by the firing rate of the photonic spiking neuron. Both experimental and numerical results show that the binary convolution is successfully achieved based on the rate-encoding properties of a single DFB-SA laser neuron. Furthermore, we numerically predict 4-channel quadratic convolution and accomplish MNIST handwritten digit classification using a spiking DFB-SA laser neuron model with rate coding. This work provides a novel approach for convolution computation, indicating the potential of integrating DFB-SA laser into future photonics spiking neural networks.

Handwritten Number Recognition Using CNN and Comparison of Performances of Models Varying by Layers

Now-a-days, the emerging technologies like Deep Learning (DL) leads a drastic turn in the area of Machine Learning (ML) by translating into Artificially Intelligent (AI). DL has several fields of applications viz. medicine, health, surveillance, drones, sports, robotics. In DL, Convolutional Neural Network (CNN) is at the centre of amazing advances that accelerates the field of visual imagery classifications. CNN has utilized broadly in face-recognition, pattern-recognition, handwritten digit-recognition, text categorization, document analysis. The objective of this work is to detect the accuracy distinction of CNN for the classification of handwritten digits via several numbers of layers and epochs. In addition to this, evaluation of the accurateness is incorporated. CNN performance was validated by Modified National Institute of Standards and Technology (MNIST) dataset. By comparing all models varying by layers, a highest validation accuracy of 99.64 % has been reported and the lowest loss gained is 0.0239.

Microdisk modulator-assisted optical nonlinear activation functions for photonic neural networks

On-chip implementation of optical nonlinear activation functions (NAFs) is essential for realizing large-scale photonic neural chips. To improve the reconfigurability and enrich the functions of photonic neural network (PNN), an NAF unit with a unique capability of performing multiple types of NAFs is highly preferred. In this work, we propose and experimentally demonstrate a microdisk modulator (MDM)-assisted NAF unit, in which multiple types of electro-optic NAFs with a high response speed and multiple types of all-optical NAFs with a low latency and reduced power consumption are performed in a single device. The fabricated MDM has an add-drop configuration, in which a lateral PN junction is incorporated to achieve high-speed resonance wavelength tuning. With the use of the high-speed nonlinear electro-optic effect, three different kinds of electro-optic NAFs including sigmoid function, radial basis function (RBF), and negative rectified linear unit (ReLU) function are realized by exploiting the free-carrier dispersion effect on silicon. Moreover, thanks to the strong optical confinement in the disk cavity, all-optical NAFs can be performed by exploiting the thermo-optic effect on silicon. In the experiment, four different kinds of all-optical NAFs including softplus function, RBF, clamped ReLU function, and leaky ReLU function are demonstrated. With the use of the realized clamped ReLU function, a convolutional neural network is simulated to perform a handwritten digit classification benchmark task, and an accuracy as high as 98 % is demonstrated. Thanks to its strong electro-optic and thermo-optic effects, the proposed MDM provides a unique capability of performing multiple types of electro-optic and all-optical NAFs, which is potential to be served as a flexible nonlinear unit in large-scale PNN chips.

A Novel Approach for Hand-written Digit Classification Using Deep Learning

A Novel Approach for Hand-written Digit Classification Using Deep Learning