Digital Commons Network^™

Cross-Lingual Transfer Learning For Statistical Type Inference, Zhiming Li, Xiaofei Xie, Haoliang Li, Zhengzi Xu, Yi Li, Yang Liu

Research Collection School Of Computing and Information Systems

Hitherto statistical type inference systems rely thoroughly on supervised learning approaches, which require laborious manual effort to collect and label large amounts of data. Most Turing-complete imperative languages share similar control- and data-flow structures, which make it possible to transfer knowledge learned from one language to another. In this paper, we propose a cross-lingual transfer learning framework, Plato, for statistical type inference, which allows us to leverage prior knowledge learned from the labeled dataset of one language and transfer it to the others, e.g., Python to JavaScript, Java to JavaScript, etc. Plato is powered by a novel kernelized attention mechanism …

Human Facial Expressions Identification Using Convolutionalneural Network With Vgg16 Architecture, Luther Alexander Latumakulita, Sandy Laurentius Lumintang, Deiby Tineke Salaki, Steven R. Sentinuwo, Alwin Melkie Sambul, Noorul Islam

Knowledge Engineering and Data Science

The human facial expression identification system is essential in developing human interaction and technology. The development of Artificial Intelligence for monitoring human emotions can be helpful in the workplace. Commonly, there are six basic human expressions, namely anger, disgust, fear, happiness, sadness, and surprise, that the system can identify. This study aims to create a facial expression identification system based on basic human expressions using the Convolutional Neural Network (CNN) with a 16-layer VGG architecture. Two thousand one hundred thirty-seven facial expression images were selected from the FER2013, JAFFE, and MUG datasets. By implementing image augmentation and setting up the …

Deep Learning Paradigm For Cardiovascular Disease/Stroke Risk Stratification In Parkinson's Disease Affected By Covid-19: A Narrative Review., Jasjit S Suri, Mahesh A Maindarkar, Sudip Paul, Puneet Ahluwalia, Mrinalini Bhagawati, Luca Saba, Gavino Faa, Sanjay Saxena, Inder M Singh, Paramjit S Chadha, Monika Turk, Amer Johri, Narendra N Khanna, Klaudija Viskovic, Sofia Mavrogeni, John R Laird, Martin Miner, David W Sobel, Antonella Balestrieri, Petros P Sfikakis, George Tsoulfas, Athanase D Protogerou, Durga Prasanna Misra, Vikas Agarwal, George D Kitas, Raghu Kolluri, Jagjit S Teji, Mustafa Al-Maini, Surinder K Dhanjil, Meyypan Sockalingam, Ajit Saxena, Aditya Sharma, Vijay Rathore, Mostafa Fatemi, Azra Alizad, Padukode R Krishnan, Tomaz Omerzu, Subbaram Naidu, Andrew Nicolaides, Kosmas I Paraskevas, Mannudeep Kalra, Zoltán Ruzsa, Mostafa M Fouda

Heart and Vascular Articles

Background and Motivation: Parkinson's disease (PD) is one of the most serious, non-curable, and expensive to treat. Recently, machine learning (ML) has shown to be able to predict cardiovascular/stroke risk in PD patients. The presence of COVID-19 causes the ML systems to become severely non-linear and poses challenges in cardiovascular/stroke risk stratification. Further, due to comorbidity, sample size constraints, and poor scientific and clinical validation techniques, there have been no well-explained ML paradigms. Deep neural networks are powerful learning machines that generalize non-linear conditions. This study presents a novel investigation of deep learning (DL) solutions for CVD/stroke risk prediction …

Machine Learning Applications To Static Timing Analysis, Waseem Mohamed Raslan

Theses and Dissertations

Modeling complex cell behavior is critical for accurate static timing analysis. Effective current source model, ECSM, and composite current source, CCS, waveform data compression became a necessity to reduce the size of technology files and increase the accuracy of the cell characterization data. We used deep learning nonlinear Autoencoders to compress voltage and current waveforms and compared them with singular value decomposition, SVD, approach. Autoencoders gave ~1.67x compression ratio for voltage waveforms better than SVD approach and gave 45x to 55x better compression ratio compared to other lossless techniques like bz2 and gzip. Autoencoders achieved ~1.7x compression ratio for complex …

Machine Learning With Kay, Lasith Niroshan, James Carswell

Conference Papers

Computational power is very important when training Deep Learning (DL) models with large amounts of data (Wooldridge, 2021). Hence, High-Performance Computing (HPC) can be leveraged to reduce computational cost, and the Irish Centre for High-End Computing (ICHEC) provides significant infrastructure and services for research and development to both academia and industry. A portion of ICHEC's HPC system has been allocated for institutional access, and this paper presents a case study of how to use Kay (Ireland's national supercomputer) in the remote sensing domain. Specifically, this study uses clusters of Kay Graphics Processing Units (GPUs) for training DL models to extract …

An Empirical Study On Sampling Approaches For 3d Image Classification Using Deep Learning, Nicholas Michelette

Theses and Dissertations

A 3D classification method requires more training data than a 2D image classification method to achieve good performance. These training data usually come in the form of multiple 2D images (e.g., slices in a CT scan) or point clouds (e.g., 3D CAD modeling) for volumetric object representation. The amount of data required to complete this higher dimension problem comes with the cost of requiring more processing time and space. This problem can be mitigated with data size reduction (i.e., sampling). In this thesis, we empirically study and compare the classification performance and deep learning training time of PointNet utilizing uniform …

Video Anomaly Detection: Practical Challenges For Learning Algorithms, Keval Doshi

USF Tampa Graduate Theses and Dissertations

Anomaly detection in surveillance videos is attracting an increasing amount of attention. Despite the competitive performance of several existing methods, they lack theoretical performance analysis, particularly due to the complex deep neural network architectures used in decision making. Additionally, real-time decision making is an important but mostly neglected factor in this domain. Much of the existing methods that claim to be online, depend on batch or offline processing in practice. Furthermore, several critical tasks such as continual learning, model interpretability and cross-domain adaptability are completely neglected in existing works. Motivated by these research gaps, in this dissertation we discuss our …

Electromyograph Estimation Of Wheelchair Operators Using Deep Learning, Shimpei Aihara, Ryusei Shibata, Ryosuke Mizukami, Takara Sakai, Akira Shionoya

International Sports Engineering Association – Engineering of Sport

No abstract provided.

Identifying Locations Of Violent Injuries In Las Vegas To Implement The Cardiff Violence Prevention Model, Samuel Vonpaays Soh, Kristi L. Herman, Chris Papesh, Tim Musgrove, Ying Zhang

SMU Data Science Review

Public violence in the United States is a major health concern. Incidents involving violent crimes are often not reported to law enforcement. The Cardiff Model is a violence prevention program developed in the UK to identify and enable data sharing of violent injury locations between Emergency Rooms (ER) and local law enforcement to help identify areas for community improvements. The model is now in use in several major cities in the US to reduce violence. Las Vegas has seen an increase in public violence in recent years. As a result, researchers from the Southern Nevada Health District (SNHD) and University …

Towards A Computational Model Of Narrative On Social Media, Anne Bailey

Dartmouth College Undergraduate Theses

This thesis describes a variety of approaches to developing a computational model of narrative on social media. Our goal is to use such a narrative model to identify efforts to manipulate public opinion on social media platforms like Twitter. We present a model in which narratives in a collection of tweets are represented as a graph. Elements from each tweet that are relevant to potential narratives are made into nodes in the graph; for this thesis, we populate graph nodes with tweets’ authors, hashtags, named entities (people, locations, organizations, etc.,), and moral foundations (central moral values framing the discussion). Two …

Training Thinner And Deeper Neural Networks: Jumpstart Regularization, Carles Riera, Camilo Rey, Thiago Serra, Eloi Puertas, Oriol Pujol

Faculty Conference Papers and Presentations

Neural networks are more expressive when they have multiple layers. In turn, conventional training methods are only successful if the depth does not lead to numerical issues such as exploding or vanishing gradients, which occur less frequently when the layers are sufficiently wide. However, increasing width to attain greater depth entails the use of heavier computational resources and leads to overparameterized models. These subsequent issues have been partially addressed by model compression methods such as quantization and pruning, some of which relying on normalization-based regularization of the loss function to make the effect of most parameters negligible. In this work, …

Deep Learning For Person Re-Identification: A Survey And Outlook, Mang Ye, Jianbing Shen, Gaojie Lin, Tao Xiang, Ling Shao, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Person re-identification (Re-ID) aims at retrieving a person of interest across multiple non-overlapping cameras. With the advancement of deep neural networks and increasing demand of intelligent video surveillance, it has gained significantly increased interest in the computer vision community. By dissecting the involved components in developing a person Re-ID system, we categorize it into the closed-world and open-world settings. We first conduct a comprehensive overview with in-depth analysis for closed-world person Re-ID from three different perspectives, including deep feature representation learning, deep metric learning and ranking optimization. With the performance saturation under closed-world setting, the research focus for person Re-ID …

Simultaneous Energy Harvesting And Gait Recognition Using Piezoelectric Energy Harvester, Dong Ma, Guohao Lan, Weitao Xu, Mahbub Hassan, Wen Hu

Research Collection School Of Computing and Information Systems

Piezoelectric energy harvester, which generates electricity from stress or vibrations, is gaining increasing attention as a viable solution to extend battery life in wearables. Recent research further reveals that, besides generating energy, PEH can also serve as a passive sensor to detect human gait power-efficiently because its stress or vibration patterns are significantly influenced by the gait. However, as PEHs are not designed for precise measurement of motion, achievable gait recognition accuracy remains low with conventional classification algorithms. The accuracy deteriorates further when the generated electricity is stored simultaneously. To classify gait reliably while simultaneously storing generated energy, we make …

Weakly-Supervised Tumor Purity Prediction From Frozen H&E Stained Slides, Matthew Brendel, Vanesa Getseva, Majd Al Assaad, Michael Sigouros, Alexandros Sigaras, Troy Kane, Pegah Khosravi, Juan Miguel Mosquera, Olivier Elemento, Iman Hajirasouliha

Publications and Research

Background

Estimating tumor purity is especially important in the age of precision medicine. Purity estimates have been shown to be critical for correction of tumor sequencing results, and higher purity samples allow for more accurate interpretations from next-generation sequencing results. Molecular-based purity estimates using computational approaches require sequencing of tumors, which is both time-consuming and expensive.

Methods

Here we propose an approach, weakly-supervised purity (wsPurity), which can accurately quantify tumor purity within a digitally captured hematoxylin and eosin (H&E) stained histological slide, using several types of cancer from The Cancer Genome Atlas (TCGA) as a proof-of-concept.

Findings

Our model predicts …

Deep Learning Modeling M6a Deposition Reveals The Importance Of Downstream Cis-Element Sequences., Zhiyuan Luo, Jiacheng Zhang, Jingyi Fei, Shengdong Ke

Faculty Research 2022

The N⁶-methyladenosine (m⁶A) modification is deposited to nascent transcripts on chromatin, but its site-specificity mechanism is mostly unknown. Here we model the m⁶A deposition to pre-mRNA by iM6A (intelligent m⁶A), a deep learning method, demonstrating that the site-specific m⁶A methylation is primarily determined by the flanking nucleotide sequences. iM6A accurately models the m⁶A deposition (AUROC = 0.99) and uncovers surprisingly that the cis-elements regulating the m⁶A deposition preferentially reside within the 50 nt downstream of the m⁶A sites. The m⁶A enhancers mostly …

Motion Field Integrated Reconstruction Using Deep Learning For Mr And Pet/Mr Respiratory Motion Correction, Sihao Chen

McKelvey School of Engineering Graduate Student Theses & Dissertations

Since motion is unavoidable in clinical human imaging studies, motion correction in PET and MRI has long been of interest for the liver/lung imaging community. Respiratory motion, the most crucial source of motion in body imaging, affects thoracic organs and the upper/lower abdomen. In MRI, motion can severely blur the images and create artifacts due to incorrect sampling of k-space data in the Fourier domain. In PET imaging, the periodic respiratory motion negatively impacts the detection of small lesions and quantification of PET tracer uptake values. Typically, breath-holding is used in clinical MR scans. However, breath-holding limits spatial coverage and …

A Machine Learning And Deep Learning Framework For Binary, Ternary, And Multiclass Emotion Classification Of Covid-19 Vaccine-Related Tweets, Aditya Dubey

Honors Scholar Theses

My research mines public emotion toward the Covid-19 vaccine based on Twitter data collected over the past 6-12 months. This project is centered around building and developing machine learning and deep learning models to perform natural language processing of short-form text, which in our case tweets. These tweets are all vaccine-related tweets and the goal of the classification task is for our models to accurately classify a tweet into one of four emotion groups: Apprehension/Anticipation, Sadness/Anger/Frustration, Joy/Humor/Sarcasm, and Gratitude/Relief. Given this data and the goal of the paper, we aim to answer the following questions: (1) Can a framework be …

Hybrid Machine And Deep Learning-Based Cyberattack Detection And Classification In Smart Grid Networks, Adedayo Aribisala

Electronic Theses and Dissertations

Power grids have rapidly evolved into Smart grids and are heavily dependent on Supervisory Control and Data Acquisition (SCADA) systems for monitoring and control. However, this evolution increases the susceptibility of the remote (VMs, VPNs) and physical interfaces (sensors, PMUs LAN, WAN, sub-stations power lines, and smart meters) to sophisticated cyberattacks. The continuous supply of power is critical to power generation plants, power grids, industrial grids, and nuclear grids; the halt to global power could have a devastating effect on the economy's critical infrastructures and human life.

Machine Learning and Deep Learning-based cyberattack detection modeling have yielded promising results when …

Improved Sensor-Based Human Activity Recognition Via Hybrid Convolutional And Recurrent Neural Networks, Sonia Perez-Gamboa

Electronic Theses, Projects, and Dissertations

Non-intrusive sensor-based human activity recognition is utilized in a spectrum of applications including fitness tracking devices, gaming, health care monitoring, and smartphone applications. Deep learning models such as convolutional neural networks (CNNs) and long short-term memory (LSTMs) recurrent neural networks provide a way to achieve human activity recognition accurately and effectively. This project designed and explored a variety of multi-layer hybrid deep learning architectures which aimed to improve human activity recognition performance by integrating local features and was scale invariant with dependencies of activities. We achieved a 94.7% activity recognition rate on the University of California, Irvine public domain dataset …

Data-Driven Design And Analysis Of Next Generation Mobile Networks For Anomaly Detection And Signal Classification With Fast, Robust And Light Machine Learning, Muhammed Furkan Küçük

USF Tampa Graduate Theses and Dissertations

This research focuses on machine (and deep) learning applications (including clustering,anomaly detection and signal classification) for self-organizing and next generation mobile networks in wireless communications. Specifically, this dissertation document will address the three different topics.

First, in the study titled “Performance analysis of neural network topologies and hyperparameters for deep clustering”, we explore the relationship between the clustering performance and network complexity. Deep learning found its initial footing in supervised applications such as image and voice recognition successes of which were followed by deep generative models across similar domains. In recent years, researchers have proposed creative learning representations to utilize …

Remote Crop Disease Detection Using Deep Learning With Iot, Ivy Chung, Anoushka Gupta

Electrical and Computer Engineering Senior Theses

Agriculture is such a vital part of our society, and according to the United Nations’ Food and Agricultural Organization (FAO), plant diseases are considered one of the two main causes of decreasing food availability. This paper explores not only the methods and findings of building a CNN-based disease detection model, but that of building a deployable remote crop disease detection system incorporating IoT technology. By using transfer learning with AlexNet, we were able to predict with 89.8% accuracy tomato plant images into one of the ten pre-defined disease classes. Our proposed system tracks plant health throughout the day by using …

When Is Deep Learning Better And When Is Shallow Learning Better: Qualitative Analysis, Salvador Robles Herrera, Martine Ceberio, Vladik Kreinovich

Departmental Technical Reports (CS)

In many practical situations, deep neural networks work better than the traditional "shallow" ones, however, in some cases, the shallow neural networks lead to better results. At present, deciding which type of neural networks will work better is mostly done by trial and error. It is therefore desirable to come up with some criterion of when deep learning is better and when shallow is better. In this paper, we argue that this depends on whether the corresponding situation has natural symmetries: if it does, we expect deep learning to work better, otherwise we expect shallow learning to be more effective. …

Quadratic Neural Network Architecture As Evaluated Relative To Conventional Neural Network Architecture, Reid Taylor

Senior Theses

Current work in the field of deep learning and neural networks revolves around several variations of the same mathematical model for associative learning. These variations, while significant and exceptionally applicable in the real world, fail to push the limits of modern computational prowess. This research does just that: by leveraging high order tensors in place of 2nd order tensors, quadratic neural networks can be developed and can allow for substantially more complex machine learning models which allow for self-interactions of collected and analyzed data. This research shows the theorization and development of mathematical model necessary for such an idea to …

Using Deep Learning And Uav Imagery To Detect Elkhorn Coral In St. Croix’S East End Marine Park, Samuel Wyatt

Master's Theses

Elkhorn coral, or Acropora palmata, is an important reef building species that promotes species abundance and other ecological services to the communities in the US Virgin Islands. We captured high resolution imagery of a reef in St. Croix’s East End Marine Park using a Wingtra One UAV. We then used deep learning techniques to detect individual coral colonies. We compared two deep learning models, FasterRCNN and MaskRCNN, and found that the models achieved accuracy shores up to 0.78. These scores improved when examining only larger corals in shallow waters. The model was able to both detect Elkhorn coral and …

Entity Based Sentiment Analysis For Textual Health Advice, Dae Lim Chung

Computer Science Senior Theses

This work explores entity based sentiment analysis for textual health advice through deep learning. We fine tuned a pretrained BERT model to analyze sentiments across five different predetermined categories which consist of food, medicine, disease, exercise, and vitality for three different sentiments: positive, negative, and neutral. Original set of annotated medical dataset from Dartmouth College’s Persist Lab was used to conduct the experiments. For the aim of tailoring the data for the purpose of entity based sentiment analysis, we explored data transformation techniques to generate optimum training examples. During the experiments, we were able to discover that the wide variety …

Deep Learning Based Generative Materials Design, Yong Zhao

Theses and Dissertations

Discovery of novel functional materials is playing an increasingly important role in many key industries such as lithium batteries for electric vehicles and cell phones. However experimental tinkering of existing materials or Density Functional Theory (DFT) based screening of known crystal structures, two of the major current materials design approaches, are both severely constrained by the limited scale (around 250,000 in ICSD database) and diversity of existing materials and the lack of a sufficient number of materials with annotated properties. How to generate a large number of physically feasible, stable, and synthesizable crystal materials and build accurate property prediction models …

A Survey On Modern Deep Neural Network For Traffic Prediction: Trends, Methods And Challenges, David Alexander Tedjopumomo, Zhifeng Bao, Baihua Zheng, Farhana Murtaza Choudhury, Kai Qin

Research Collection School Of Computing and Information Systems

In this modern era, traffic congestion has become a major source of negative economic and environmental impact for urban areas worldwide. One of the most efficient ways to mitigate traffic congestion is through future traffic prediction. The field of traffic prediction has evolved greatly ever since its inception in the late 70s. Earlier studies mainly use classical statistical models such as ARIMA and its variants. Then, researchers started to focus on machine learning models due to their power and flexibility. As theoretical and technological advances emerge, we enter the era of deep neural network, which gained popularity due to its …

Fusionai, A Dna-Sequence-Based Deep Learning Protocol Reduces The False Positives Of Human Fusion Gene Prediction, Xiaobo Zhou

Faculty, Staff and Student Publications

Even though there were many tool developments of fusion gene prediction from NGS data, too many false positives are still an issue. Wise use of the genomic features around the fusion gene breakpoints will be helpful to identify reliable fusion genes efficiently. For this aim, we developed FusionAI, a deep learning pipeline predicting human fusion gene breakpoints from DNA sequence. FusionAI is freely available via https://compbio.uth.edu/FusionGDB2/FusionAI. For complete details on the use and execution of this protocol, please refer to Kim et al. (2021b).

Riconv++: Effective Rotation Invariant Convolutions For 3d Point Clouds Deep Learning, Zhiyuan Zhang, Binh-Son Hua, Sai-Kit Yeung

Research Collection School Of Computing and Information Systems

3D point clouds deep learning is a promising field of research that allows a neural network to learn features of point clouds directly, making it a robust tool for solving 3D scene understanding tasks. While recent works show that point cloud convolutions can be invariant to translation and point permutation, investigations of the rotation invariance property for point cloud convolution has been so far scarce. Some existing methods perform point cloud convolutions with rotation-invariant features, existing methods generally do not perform as well as translation-invariant only counterpart. In this work, we argue that a key reason is that compared to …

Use Of The Deep Learning Approach To Measure Alveolar Bone Level, Hsiu-Wan Meng, Thomas E Van Dyke, Muhammad F Walji, Xiaoqian Jiang, Shayan Shams

Faculty, Staff and Student Publications

AIM: The goal was to use a deep convolutional neural network to measure the radiographic alveolar bone level to aid periodontal diagnosis.

MATERIALS AND METHODS: A deep learning (DL) model was developed by integrating three segmentation networks (bone area, tooth, cemento-enamel junction) and image analysis to measure the radiographic bone level and assign radiographic bone loss (RBL) stages. The percentage of RBL was calculated to determine the stage of RBL for each tooth. A provisional periodontal diagnosis was assigned using the 2018 periodontitis classification. RBL percentage, staging, and presumptive diagnosis were compared with the measurements and diagnoses made by the …