Md. Zahangir Alom

Memristor crossbar deep network implementation based on a Convolutional neural network

This paper presents a simulated memristor crossbar implementation of a deep Convolutional Neural Network (CNN). In the past few years deep neural networks implemented on GPU clusters have become the state of the art in image classification. They provide excellent classification ability at the cost of a more complex data manipulation process. However once these systems are trained, we show that the analog crossbar circuits in this paper can highly parallelize the recognition phase of a CNN algorithm. One of the drawbacks of using memristors to carry out computations is that the data stored will likely have less precision when compared to typical 32-bit floating point memory. However, we show the proposed system is capable of operating with zero loss in classification accuracy if the memristors utilized are able to store at least 16 unique values (essentially acting as 4-bit devices). To the best of our knowledge, this is the first paper that presents a memristor based circuit for implementing CNN recognition. This is also the first paper that provides a circuit for precise memristor based analog convolution.

Intrusion detection using deep belief networks

With the advent of digital technology, security threats for computer networks have increased dramatically over the last decade being much bolder and brazen. There is a great need for an effective Intrusion Detection System (IDS) which are intelligent specialized system designed to interpret the intrusion attempts in incoming network traffic. Deep belief neural (DBN) networks proved to be the most influential deep neural nets and generative neural networks that stack Restricted Boltzmann Machines. In this paper, we explore the capabilities of DBNs performing intrusion detection through series of experiments after training it with NSL-KDD dataset. The trained DBN network now identifies any kind of unknown attack in dataset supplied to it and to the best of our knowledge this is first comprehensive paper performing intrusion detection using deep belief nets. The proposed system not only detect attacks but also classify them in five groups with the accuracy of identifying and classifying network activity based on limited, incomplete, and nonlinear data sources. The proposed system achieved detection accuracy about 97.5% for only fifty iterations that is state of art performance compare to the existing system till today for intrusion detection.

State Preserving Extreme Learning Machine for face recognition

Extreme Learning Machine (ELM) has been introduced as a new algorithm for training single hidden layer feed-forward neural networks (SLFNs) instead of the classical gradient-based algorithms. Based on the consistency property of data, which enforce similar samples to share similar properties, ELM is a biologically inspired learning algorithm with SLFNs that learns much faster with good generalization and performs well in classification applications. However, the random generation of the weight matrix in current ELM based techniques leads to the possibility of unstable outputs in the learning and testing phases. Therefore, we present a novel approach for computing the weight matrix in ELM which forms a State Preserving Extreme Leaning Machine (SPELM). The SPELM stabilizes ELM training and testing outputs while monotonically increases its accuracy by preserving state variables. Furthermore, three popular feature extraction techniques, namely Gabor, Pyramid Histogram of Oriented Gradients (PHOG) and Local Binary Pattern (LBP) are incorporated with the SPELM for performance evaluation. Experimental results show that our proposed algorithm yields the best performance on the widely used face datasets such as Yale, CMU and ORL compared to state-of-the-art ELM based classifiers.

State Preserving Extreme Learning Machine: A Monotonically Increasing Learning Approach

Extreme Learning Machines (ELM) has been introduced as a new algorithm for training single hidden layer feedforward neural networks instead of the classical gradient-based approaches. Based on the consistency property of data, which enforces similar samples to share similar properties, ELM is a biologically inspired learning algorithm that learns much faster with good generalization and performs well in classification tasks. However, the stochastic characteristics of hidden layer outputs from the random generation of the weight matrix in current ELMs leads to the possibility of unstable outputs in the learning and testing phases. This is detrimental to the overall performance when many repeated trials are conducted. To cope with this issue, we present a new ELM approach, named State Preserving Extreme Leaning Machine (SPELM). SPELM ensures the overall training and testing performance of the classical ELM while monotonically increases its accuracy by preserving state variables. For evaluation, experiments are performed on different benchmark datasets including applications in face recognition, pedestrian detection, and network intrusion detection for cyber security. Several popular feature extraction techniques, namely Gabor, pyramid histogram of oriented gradients, and local binary pattern are also incorporated with SPELM. Experimental results show that our SPELM algorithm yields the best performance on tested data over ELM and RELM.

Optical beam classification using deep learning: a comparison with rule- and feature-based classification

Deep-learning methods are gaining popularity because of their state-of-the-art performance in image classification tasks. In this paper, we explore classification of laser-beam images from the National Ignition Facility (NIF) using a novel deeplearning approach. NIF is the world’s largest, most energetic laser. It has nearly 40,000 optics that precisely guide, reflect, amplify, and focus 192 laser beams onto a fusion target. NIF utilizes four petawatt lasers called the Advanced Radiographic Capability (ARC) to produce backlighting X-ray illumination to capture implosion dynamics of NIF experiments with picosecond temporal resolution. In the current operational configuration, four independent short-pulse ARC beams are created and combined in a split-beam configuration in each of two NIF apertures at the entry of the pre-amplifier. The subaperture beams then propagate through the NIF beampath up to the ARC compressor. Each ARC beamlet is separately compressed with a dedicated set of four gratings and recombined as sub-apertures for transport to the parabola vessel, where the beams are focused using parabolic mirrors and pointed to the target. Small angular errors in the compressor gratings can cause the sub-aperture beams to diverge from one another and prevent accurate alignment through the transport section between the compressor and parabolic mirrors. This is an off-normal condition that must be detected and corrected. The goal of the off-normal check is to determine whether the ARC beamlets are sufficiently overlapped into a merged single spot or diverged into two distinct spots. Thus, the objective of the current work is three-fold: developing a simple algorithm to perform off-normal classification, exploring the use of Convolutional Neural Network (CNN) for the same task, and understanding the inter-relationship of the two approaches. The CNN recognition results are compared with other machine-learning approaches, such as Deep Neural Network (DNN) and Support Vector Machine (SVM). The experimental results show around 96% classification accuracy using CNN; the CNN approach also provides comparable recognition results compared to the present feature-based off-normal detection. The feature-based solution was developed to capture the expertise of a human expert in classifying the images. The misclassified results are further studied to explain the differences and discover any discrepancies or inconsistencies in current classification.

Non-regularized State Preserving Extreme Learning Machine for Natural Scene Classification

Scene classification remains a challenging task in computer vision applications due to a wide range of intraclass and interclass variations. A robust feature extraction technique and an effective classifier are required to achieve satisfactory recognition performance. Herein, we propose a nonregularized state preserving extreme learning machine (NSPELM) to perform scene classification tasks. We employ a Bag-of-Words (BoW) model for feature extraction prior to performing the classification task. The BoW feature is obtained based on a regular grid method for point selection and Speeded Up Robust Features (SURF) technique for feature extraction on the selected points. The performance of NSPELM is tested and evaluated on three standard scene category classification datasets. The recognition accuracy is compared with the standard extreme learning machine classifier and it shows that the proposed NSPELM algorithm yields better accuracy.