Computer Sciences Commons^™

Channel-Resilient Deep-Learning-Driven Device Fingerprinting Through Multiple Data Streams, Nora Basha, Bechir Hamdaoui, Kathiravetpillai Sivanesan, Mohsen Guizani

Machine Learning Faculty Publications

Enabling accurate and automated identification of wireless devices is critical for allowing network access monitoring and ensuring data authentication for large-scale IoT networks. RF fingerprinting has emerged as a solution for device identification by leveraging the transmitters' inevitable hardware impairments that occur during manufacturing. Although deep learning is proven efficient in classifying devices based on hardware impairments, the performance of deep learning models suffers greatly from variations of the wireless channel conditions, across time and space. To the best of our knowledge, we are the first to propose leveraging MIMO capabilities to mitigate the channel effect and provide a channel-resilient …

Encryption And Compression Classification Of Internet Of Things Traffic, Mariam Najdat M Saleh

Browse all Theses and Dissertations

The Internet of Things (IoT) is used in many fields that generate sensitive data, such as healthcare and surveillance. Increased reliance on IoT raised serious information security concerns. This dissertation presents three systems for analyzing and classifying IoT traffic using Deep Learning (DL) models, and a large dataset is built for systems training and evaluation. The first system studies the effect of combining raw data and engineered features to optimize the classification of encrypted and compressed IoT traffic using Engineered Features Classification (EFC), Raw Data Classification (RDC), and combined Raw Data and Engineered Features Classification (RDEFC) approaches. Our results demonstrate …

Nudyclr: Nuclear Dynamic Co-Learned Representations, Víctor Samuel Pérez-Díaz

2023 REYES Proceedings

NuCLR (Nuclear Co-Learned Representations) is a cutting-edge multi-task deep learning framework designed to predict essential nuclear observables, including binding energies, decay energies, and nuclear charge radii. As part of the REYES Mentorship Program, we investigated the application of dynamic loss weighting to further refine NuCLR’s predictive performance. Our findings indicate that while weighting strategies can enhance accuracy in specific tasks, such as binding energy prediction, they may underperform in others. Equal Weighting (EW), the original method employed by NuCLR, demonstrated consistent performance across multiple tasks, affirming its robustness. This report succinctly presents the developments and results of the mentorship program …

Detecting Patches On Road Pavement Images Acquired With 3d Laser Sensors Using Object Detection And Deep Learning, Ibrahim Hassan Syed, Dympna O'Sullivan, Susan Mckeever, David Power, Ray Mcgowan, Kieran Feighan

Academic Posters Collection

Regular pavement inspections are key to good road maintenance and road defect corrections. Advanced pavement inspection systems such as LCMS (Laser Crack Measurement System) can automatically detect the presence of different defects using 3D lasers. However, such systems still require manual involvement to complete the detection of pavement defects. This work proposes an automatic patch detection system using an object detection technique. Results show that the object detection model can successfully detect patches inside LCMS images and suggest that the proposed approach could be integrated into the existing pavement inspection systems.

A Multistage Framework For Detection Of Very Small Objects, Duleep Rathgamage Don, Ramazan Aygun, Mahmut Karakaya

Published and Grey Literature from PhD Candidates

Small object detection is one of the most challenging problems in computer vision. Algorithms based on state-of-the-art object detection methods such as R-CNN, SSD, FPN, and YOLO fail to detect objects of very small sizes. In this study, we propose a novel method to detect very small objects, smaller than 8×8 pixels, that appear in a complex background. The proposed method is a multistage framework consisting of an unsupervised algorithm and three separately trained supervised algorithms. The unsupervised algorithm extracts ROIs from a high-resolution image. Then the ROIs are upsampled using SRGAN, and the enhanced ROIs are detected by our …

A Deep Bilstm Machine Learning Method For Flight Delay Prediction Classification, Desmond B. Bisandu, Irene Moulitsas

Journal of Aviation/Aerospace Education & Research

This paper proposes a classification approach for flight delays using Bidirectional Long Short-Term Memory (BiLSTM) and Long Short-Term Memory (LSTM) models. Flight delays are a major issue in the airline industry, causing inconvenience to passengers and financial losses to airlines. The BiLSTM and LSTM models, powerful deep learning techniques, have shown promising results in a classification task. In this study, we collected a dataset from the United States (US) Bureau of Transportation Statistics (BTS) of flight on-time performance information and used it to train and test the BiLSTM and LSTM models. We set three criteria for selecting highly important features …

Explaining Deep Learning Time Series Classification Models Using A Decision Tree, Ephrem T. Mekonnen, Pierpaolo Dondio, Luca Longo

Academic Posters Collection

This preliminary study proposes a new post hoc method to explain deep learning-based time series classification models using a decision tree. Our approach generates a decision tree graph or rulesets as an explanation, improving interpretability compared to saliency map-based methods. The method involves two phases: training and evaluating the deep learning-based time series classification model and extracting prototypical events from the evaluation set to train the decision tree classifier. We conducted experiments on artificial and real datasets, evaluating the explanations based on accuracy, fidelity, number of nodes, and depth. Our preliminary findings suggest that our post-hoc method improves the interpretability …

Communety: Deep Learning-Based Face Recognition System For The Prediction Of Cohesive Communities, Syed Afaq Ali Shah, Weifeng Deng, Muhammad Aamir Cheema, Abdul Bais

Research outputs 2022 to 2026

Effective mining of social media, which consists of a large number of users is a challenging task. Traditional approaches rely on the analysis of text data related to users to accomplish this task. However, text data lacks significant information about the social users and their associated groups. In this paper, we propose CommuNety, a deep learning system for the prediction of cohesive networks using face images from photo albums. The proposed deep learning model consists of hierarchical CNN architecture to learn descriptive features related to each cohesive network. The paper also proposes a novel Face Co-occurrence Frequency algorithm to quantify …

Causal Interventional Training For Image Recognition, Wei Qin, Hanwang Zhang, Richang Hong, Ee-Peng Lim, Qianru Sun

Research Collection School Of Computing and Information Systems

Deep learning models often fit undesired dataset bias in training. In this paper, we formulate the bias using causal inference, which helps us uncover the ever-elusive causalities among the key factors in training, and thus pursue the desired causal effect without the bias. We start from revisiting the process of building a visual recognition system, and then propose a structural causal model (SCM) for the key variables involved in dataset collection and recognition model: object, common sense, bias, context, and label prediction. Based on the SCM, one can observe that there are “good” and “bad” biases. Intuitively, in the image …

Adversarial Training Of Deep Neural Networks, Anabetsy Termini

CCAC Theses and Dissertations

Deep neural networks used for image classification are highly susceptible to adversarial attacks. The de facto method to increase adversarial robustness is to train neural networks with a mixture of adversarial images and unperturbed images. However, this method leads to robust overfitting, where the network primarily learns to recognize one specific type of attack used to generate the images while remaining vulnerable to others after training. In this dissertation, we performed a rigorous study to understand whether combinations of state of the art data augmentation methods with Stochastic Weight Averaging improve adversarial robustness and diminish adversarial overfitting across a wide …

Mobilenetv1-Based Deep Learning Model For Accurate Brain Tumor Classification, Maad M. Mijwil, Ruchi Doshi, Kamal Kant Hiran, Omega John Unogwu, Indu Bala

Mesopotamian Journal of Computer Science

Brain tumors are among the most dangerous diseases that lead to mortality after a period of time from injury. Therefore, physicians and healthcare professionals are advised to make an early diagnosis of brain tumors and follow their instructions. Magnetic resonance imaging (MRI) is operated to provide sufficient and practical data in detecting brain tumors. Applications based on artificial intelligence contribute a very large role in disease detection, provide incredible accuracy and assist in creating the right decisions. In particular, deep learning models, which are a significant part of artificial intelligence, have the ability to diagnose and process medical image datasets. …

Apt Adversarial Defence Mechanism For Industrial Iot Enabled Cyber-Physical System, Safdar Hussain Javed, Maaz Bin Ahmad, Muhammad Asif, Waseem Akram, Khalid Mahmood, Ashok Kumar Das, Sachin Shetty

VMASC Publications

The objective of Advanced Persistent Threat (APT) attacks is to exploit Cyber-Physical Systems (CPSs) in combination with the Industrial Internet of Things (I-IoT) by using fast attack methods. Machine learning (ML) techniques have shown potential in identifying APT attacks in autonomous and malware detection systems. However, detecting hidden APT attacks in the I-IoT-enabled CPS domain and achieving real-time accuracy in detection present significant challenges for these techniques. To overcome these issues, a new approach is suggested that is based on the Graph Attention Network (GAN), a multi-dimensional algorithm that captures behavioral features along with the relevant information that other methods …

Comparative Adjudication Of Noisy And Subjective Data Annotation Disagreements For Deep Learning, Scott David Williams

Browse all Theses and Dissertations

Obtaining accurate inferences from deep neural networks is difficult when models are trained on instances with conflicting labels. Algorithmic recognition of online hate speech illustrates this. No human annotator is perfectly reliable, so multiple annotators evaluate and label online posts in a corpus. Labeling scheme limitations, differences in annotators' beliefs, and limits to annotators' honesty and carefulness cause some labels to disagree. Consequently, decisive and accurate inferences become less likely. Some practical applications such as social research can tolerate some indecisiveness. However, an online platform using an indecisive classifier for automated content moderation could create more problems than it solves. …

An Empirical Study Of Pre-Trained Model Reuse In The Hugging Face Deep Learning Model Registry, Wenxin Jiang, Nicholas Synovic, Matt Hyatt, Taylor R. Schorlemmer, Rohan Sethi, Yung-Hsiang Lu, George K. Thiruvathukal, James C. Davis

Department of Electrical and Computer Engineering Faculty Publications

Deep Neural Networks (DNNs) are being adopted as components in software systems. Creating and specializing DNNs from scratch has grown increasingly difficult as state-of-the-art architectures grow more complex. Following the path of traditional software engineering, machine learning engineers have begun to reuse large-scale pre-trained models (PTMs) and fine-tune these models for downstream tasks. Prior works have studied reuse practices for traditional software packages to guide software engineers towards better package maintenance and dependency management. We lack a similar foundation of knowledge to guide behaviors in pre-trained model ecosystems.

In this work, we present the first empirical investigation of PTM reuse. …

Deep Learning For Medical Image Segmentation Using Prior Knowledge And Topology, Chen Zhao

Dissertations, Master's Theses and Master's Reports

Image segmentation refers to the division of a digital image into distinct segments or groups of pixels/voxels. However, most of the existing deep learning approaches lack the utilization of prior knowledge, such as shape information, which could improve segmentation accuracy. In addition, conventional image segmentation frequently falls short in preserving intricate spatial details, motivating the innovation of strategies for multi-scaled feature integration. Furthermore, traditional image segmentation methods primarily concentrate on pixel-level or region-level analysis. However, given the inherent morphological similarities among various image objects, the significance of topology information surpasses that of pixel-level data in the realm of medical image …

Multi-Label Text Classification With Transfer Learning, Likhitha Yelamanchili

Master's Projects

Multi-label text categorization is a crucial task in Natural Language Processing, where each text instance can be simultaneously assigned to numerous labels. This project's goal is to assess how well several deep learning models perform on a real-world dataset for multi-label text classification. We employed data augmentation techniques like Synonym Substitution and Random Word Substitution to address the problem of data imbalance. We conducted experiments on a toxic comment classification dataset to evaluate the effectiveness of several deep learning models including Bi-LSTM, GRU, and Bi-GRU, as well as fine- tuned pre-trained BERT models. Many metrics, including log loss, recall@k, and …

An Explainable Deep Learning Prediction Model For Severity Of Alzheimer's Disease From Brain Images, Godwin O. Ekuma

Graduate Theses/Dissertations

Deep Convolutional Neural Networks (CNNs) have become the go-to method for medical imaging classification on various imaging modalities for binary and multiclass problems. Deep CNNs extract spatial features from image data hierarchically, with deeper layers learning more relevant features for the classification application. The effectiveness of deep learning models are hampered by limited data sets, skewed class distributions, and the undesirable "black box" of neural networks, which decreases their understandability and usability in precision medicine applications. This thesis addresses the challenge of building an explainable deep learning model for a clinical application: predicting the severity of Alzheimer's disease (AD). AD …

Learning Large Neighborhood Search For Vehicle Routing In Airport Ground Handling, Jianan Zhou, Yaoxin Wu, Zhiguang Cao, Wen Song, Jie Zhang, Zhenghua Chen

Research Collection School Of Computing and Information Systems

Dispatching vehicle fleets to serve flights is a key task in airport ground handling (AGH). Due to the notable growth of flights, it is challenging to simultaneously schedule multiple types of operations (services) for a large number of flights, where each type of operation is performed by one specific vehicle fleet. To tackle this issue, we first represent the operation scheduling as a complex vehicle routing problem and formulate it as a mixed integer linear programming (MILP) model. Then given the graph representation of the MILP model, we propose a learning assisted large neighborhood search (LNS) method using data generated …

View Synthesis With Scene Recognition For Cross-View Image Localization, Uddom Lee, Peng Jiang, Hongyi Wu, Chunsheng Xin

Electrical & Computer Engineering Faculty Publications

Image-based localization has been widely used for autonomous vehicles, robotics, augmented reality, etc., and this is carried out by matching a query image taken from a cell phone or vehicle dashcam to a large scale of geo-tagged reference images, such as satellite/aerial images or Google Street Views. However, the problem remains challenging due to the inconsistency between the query images and the large-scale reference datasets regarding various light and weather conditions. To tackle this issue, this work proposes a novel view synthesis framework equipped with deep generative models, which can merge the unique features from the outdated reference dataset with …

Security Of Internet Of Things (Iot) Using Federated Learning And Deep Learning — Recent Advancements, Issues And Prospects, Vinay Gugueoth, Sunitha Safavat, Sachin Shetty

Electrical & Computer Engineering Faculty Publications

There is a great demand for an efficient security framework which can secure IoT systems from potential adversarial attacks. However, it is challenging to design a suitable security model for IoT considering the dynamic and distributed nature of IoT. This motivates the researchers to focus more on investigating the role of machine learning (ML) in the designing of security models. A brief analysis of different ML algorithms for IoT security is discussed along with the advantages and limitations of ML algorithms. Existing studies state that ML algorithms suffer from the problem of high computational overhead and risk of privacy leakage. …

Deep Learning-Based Classification Of Chaotic Systems Over Phase Portraits, Sezgi̇n Kaçar, Süleyman Uzun, Burak Aricioğlu

Turkish Journal of Electrical Engineering and Computer Sciences

This study performed a deep learning-based classification of chaotic systems over their phase portraits. To the best of the authors' knowledge, such classification studies over phase portraits have not been conducted in the literature. To that end, a dataset consisting of the phase portraits of the most known two chaotic systems, namely Lorenz and Chen, is generated for different values of the parameters, initial conditions, step size, and time length. Then, a classification with high accuracy is carried out employing transfer learning methods. The transfer learning methods used in the study are SqueezeNet, VGG-19, AlexNet, ResNet50, ResNet101, DenseNet201, ShuffleNet, and …

Innovations In Drop Shape Analysis Using Deep Learning And Solving The Young-Laplace Equation For An Axisymmetric Pendant Drop, Andres P. Hyer

Theses and Dissertations

Axisymmetric Drop Shape Analysis (ADSA) is a technique commonly used to determine surface or interfacial tension. Applications of traditional ASDA methods to process analytical technologies are limited by computational speed and image quality. Here, we address these limitations using a novel machine learning approach to analysis. With a convolutional neural network (CNN), we were able to achieve an experimental fit precision of (+/-) 0.122 mN/m in predicting the surface tension of drop images at a rate of 1.5 ms^-1 versus 7.7 s^-1, which is more than 5,000 times faster than the traditional method. The results are validated on real images …

Face Anti-Spoofing And Deep Learning Based Unsupervised Image Recognition Systems, Enoch Solomon

Theses and Dissertations

One of the main problems of a supervised deep learning approach is that it requires large amounts of labeled training data, which are not always easily available. This PhD dissertation addresses the above-mentioned problem by using a novel unsupervised deep learning face verification system called UFace, that does not require labeled training data as it automatically, in an unsupervised way, generates training data from even a relatively small size of data. The method starts by selecting, in unsupervised way, k-most similar and k-most dissimilar images for a given face image. Moreover, this PhD dissertation proposes a new loss function to …

Facial Expression Recognition Using Lightweight Deep Learning Modeling, Mubashir Ahmad, Saira, Omar Alfandi, Asad Masood Khattak, Syed Furqan Qadri, Iftikhar Ahmed Saeed, Salabat Khan, Bashir Hayat, Arshad Ahmad

All Works

Facial expression is a type of communication and is useful in many areas of computer vision, including intelligent visual surveillance, human-robot interaction and human behavior analysis. A deep learning approach is presented to classify happy, sad, angry, fearful, contemptuous, surprised and disgusted expressions. Accurate detection and classification of human facial expression is a critical task in image processing due to the inconsistencies amid the complexity, including change in illumination, occlusion, noise and the over-fitting problem. A stacked sparse auto-encoder for facial expression recognition (SSAE-FER) is used for unsupervised pre-training and supervised fine-tuning. SSAE-FER automatically extracts features from input images, and …

Multimodal Neuron Classification Based On Morphology And Electrophysiology, Aqib Ahmad

Graduate Theses, Dissertations, and Problem Reports (ETD)

Categorizing neurons into different types to understand neural circuits and ultimately brain function is a major challenge in neuroscience. While electrical properties are critical in defining a neuron, its morphology is equally important. Advancements in single-cell analysis methods have allowed neuroscientists to simultaneously capture multiple data modalities from a neuron. We propose a method to classify neurons using both morphological structure and electrophysiology. Current approaches are based on a limited analysis of morphological features. We propose to use a new graph neural network to learn representations that more comprehensively account for the complexity of the shape of neuronal structures. In …

Social Media Bot Detection With Deep Learning Methods: A Systematic Review, Kadhim Hayawi, Susmita Saha, Mohammad Mehedy Masud, Sujith Samuel Mathew, Mohammed Kaosar

All Works

Social bots are automated social media accounts governed by software and controlled by humans at the backend. Some bots have good purposes, such as automatically posting information about news and even to provide help during emergencies. Nevertheless, bots have also been used for malicious purposes, such as for posting fake news or rumour spreading or manipulating political campaigns. There are existing mechanisms that allow for detection and removal of malicious bots automatically. However, the bot landscape changes as the bot creators use more sophisticated methods to avoid being detected. Therefore, new mechanisms for discerning between legitimate and bot accounts are …

A Survey On Artificial Intelligence-Based Acoustic Source Identification, Ruba Zaheer, Iftekhar Ahmad, Daryoush Habibi, Kazi Y. Islam, Quoc Viet Phung

Research outputs 2022 to 2026

The concept of Acoustic Source Identification (ASI), which refers to the process of identifying noise sources has attracted increasing attention in recent years. The ASI technology can be used for surveillance, monitoring, and maintenance applications in a wide range of sectors, such as defence, manufacturing, healthcare, and agriculture. Acoustic signature analysis and pattern recognition remain the core technologies for noise source identification. Manual identification of acoustic signatures, however, has become increasingly challenging as dataset sizes grow. As a result, the use of Artificial Intelligence (AI) techniques for identifying noise sources has become increasingly relevant and useful. In this paper, we …

Cooperative Deep Q -Learning Framework For Environments Providing Image Feedback, Krishnan Raghavan, Vignesh Narayanan, Sarangapani Jagannathan

Publications

In this article, we address two key challenges in deep reinforcement learning (DRL) setting, sample inefficiency, and slow learning, with a dual-neural network (NN)-driven learning approach. In the proposed approach, we use two deep NNs with independent initialization to robustly approximate the action-value function in the presence of image inputs. In particular, we develop a temporal difference (TD) error-driven learning (EDL) approach, where we introduce a set of linear transformations of the TD error to directly update the parameters of each layer in the deep NN. We demonstrate theoretically that the cost minimized by the EDL regime is an approximation …

Mwirgan: Unsupervised Visible-To Mwir Image Translation With Generative Adversarial Network, Mohammad Shahab Uddin, Chiman Kwan, Jiang Li

Electrical & Computer Engineering Faculty Publications

Unsupervised image-to-image translation techniques have been used in many applications, including visible-to-Long-Wave Infrared (visible-to-LWIR) image translation, but very few papers have explored visible-to-Mid-Wave Infrared (visible-to-MWIR) image translation. In this paper, we investigated unsupervised visible-to-MWIR image translation using generative adversarial networks (GANs). We proposed a new model named MWIRGAN for visible-to-MWIR image translation in a fully unsupervised manner. We utilized a perceptual loss to leverage shape identification and location changes of the objects in the translation. The experimental results showed that MWIRGAN was capable of visible-to-MWIR image translation while preserving the object’s shape with proper enhancement in the translated images and …

Toward Real-Time, Robust Wearable Sensor Fall Detection Using Deep Learning Methods: A Feasibility Study, Haben Yhdego, Christopher Paolini, Michel Audette

Electrical & Computer Engineering Faculty Publications

Real-time fall detection using a wearable sensor remains a challenging problem due to high gait variability. Furthermore, finding the type of sensor to use and the optimal location of the sensors are also essential factors for real-time fall-detection systems. This work presents real-time fall-detection methods using deep learning models. Early detection of falls, followed by pneumatic protection, is one of the most effective means of ensuring the safety of the elderly. First, we developed and compared different data-segmentation techniques for sliding windows. Next, we implemented various techniques to balance the datasets because collecting fall datasets in the real-time setting has …