Physical Sciences and Mathematics Commons^™

Sparse Representation Learning For Temporal Networks, Maxwell Mcneil

Electronic Theses & Dissertations (2024 - present)

Temporal networks arise in many domains including activity of social network users, sensor network readings over time, and time course gene expression within the interaction network of a model organism. Data of this type contains a wealth of prior information such as the connectivity among nodes (e.g., a friendship graph), and prior knowledge of expected temporal patterns (e.g., periodicity). Modeling these temporal and network patterns jointly is essential for state-of-the-art performance in temporal network data analysis and mining. Sparse dictionary encoding is one modeling approach for such underlying patterns. However, most classical approaches consider only one dimension of the data …

Towards Algorithmic Justice: Human Centered Approaches To Artificial Intelligence Design To Support Fairness And Mitigate Bias In The Financial Services Sector, Jihyun Kim

CMC Senior Theses

Artificial Intelligence (AI) has positively transformed the Financial services sector but also introduced AI biases against protected groups, amplifying existing prejudices against marginalized communities. The financial decisions made by biased algorithms could cause life-changing ramifications in applications such as lending and credit scoring. Human Centered AI (HCAI) is an emerging concept where AI systems seek to augment, not replace human abilities while preserving human control to ensure transparency, equity and privacy. The evolving field of HCAI shares a common ground with and can be enhanced by the Human Centered Design principles in that they both put humans, the user, at …

The Integration Of Neuromorphic Computing In Autonomous Robotic Systems, Md Abu Bakr Siddique

Dissertations, Master's Theses and Master's Reports

Deep Neural Networks (DNNs) have come a long way in many cognitive tasks by training on large, labeled datasets. However, this method has problems in places with limited data and energy, like when planetary robots are used or when edge computing is used [1]. In contrast to this data-heavy approach, animals demonstrate an innate ability to learn by communicating with their environment and forming associative memories among events and entities, a process known as associative learning [2-4]. For instance, rats in a T-maze learn to associate different stimuli with outcomes through exploration without needing labeled data [5]. This learning paradigm …

Applications Of Ai/Ml In Maritime Cyber Supply Chains, Rafael Diaz, Ricardo Ungo, Katie Smith, Lida Haghnegahdar, Bikash Singh, Tran Phuong

School of Cybersecurity Faculty Publications

Digital transformation is a new trend that describes enterprise efforts in transitioning manual and likely outdated processes and activities to digital formats dominated by the extensive use of Industry 4.0 elements, including the pervasive use of cyber-physical systems to increase efficiency, reduce waste, and increase responsiveness. A new domain that intersects supply chain management and cybersecurity emerges as many processes as possible of the enterprise require the convergence and synchronizing of resources and information flows in data-driven environments to support planning and execution activities. Protecting the information becomes imperative as big data flows must be parsed and translated into actions …

Dark Side Of Genai: A Blackbox Analysis Of X, Ahmed El Noshokaty, Tareq Nasralah, Omar El-Gayar, Mohammad A. Al-Ramahi, Abdullah Wahbeh

All Faculty Scholarship

Recent advancements in generative artificial intelligence (GenAI) have raised many fears, risks, and concerns (Kim 2023; Okey et al. 2023). To shed light on the dark side of GenAI, we collected 55,916 posts from X (formerly Twitter). Based on the content of these posts, we manually labeled a sample set with the corresponding dark side, then identified a short, comprehensive list of GenAI dark sides. Using this list, we trained the ReadMe classifier, a supervised learning algorithm on Brandwatch (“Crimson Hexagon and Brandwatch” 2020), to classify the remaining posts. Further analysis, including emotion analysis and analysis of professions and interests …

Computer-Aided Craniofacial Superimposition Validation Study: The Identification Of The Leaders And Participants Of The Polish-Lithuanian January Uprising (1863–1864), Rubén Martos, Rosario Guerra, Fernando Navarro, Michela Peruch, Kevin Neuwirth, Andrea Valsecchi, Rimantas Jankauskas, Oscar Ibáñez

Student and Faculty Publications

In 2017, a series of human remains corresponding to the executed leaders of the "January Uprising" of 1863-1864 were uncovered at the Upper Castle of Vilnius (Lithuania). During the archeological excavations, 14 inhumation pits with the human remains of 21 individuals were found at the site. The subsequent identification process was carried out, including the analysis and cross-comparison of post-mortem data obtained in situ and in the lab with ante-mortem data obtained from historical archives. In parallel, three anthropologists with diverse backgrounds in craniofacial identification and two students without previous experience attempted to identify 11 of these 21 individuals using …

Robot-Based 3d Printing, Aaron Hoffman

Williams Honors College, Honors Research Projects

Details of a large-format 3D printer created to print experimental materials, test multi-axis print techniques, and quickly print large objects. The printer consists of a 7-axis robotic arm and pellet extruder, which are controlled by a PC. Experimental materials such as recycled polymers or carbon-fiber reinforced materials can be easily tested with the pellet format of the extruder. The printer can perform different printing techniques and can be used to experiment with material properties when using these techniques with different polymers. The print surface is around 5 times larger than the average commercial 3D printer, and the robotic arm provides …

A Comparison Of Lexical Tokenization Methods, Nathan Culmer

Williams Honors College, Honors Research Projects

The purpose of this project was to compare tokenization methods, or methods of breaking up a text into meaningful parts for use in natural language processing. The effectiveness of several commonly used tokenization methods were investigated, including morpheme tokenization, which takes into account the linguistic features of the language. In addition, I proposed and implemented a new technique to consider the capitalization pattern of a word in the tokenization process, in order to allow this process to include more natural language features. The effectiveness of these methods was compared by using them in a sentiment analysis model for various datasets, …

A Review Of Hybrid Cyber Threats Modelling And Detection Using Artificial Intelligence In Iiot, Yifan Liu, Shancang Li, Xinheng Wang, Li Xu

Information Technology & Decision Sciences Faculty Publications

The Industrial Internet of Things (IIoT) has brought numerous benefits, such as improved efficiency, smart analytics, and increased automation. However, it also exposes connected devices, users, applications, and data generated to cyber security threats that need to be addressed. This work investigates hybrid cyber threats (HCTs), which are now working on an entirely new level with the increasingly adopted IIoT. This work focuses on emerging methods to model, detect, and defend against hybrid cyber attacks using machine learning (ML) techniques. Specifically, a novel ML-based HCT modelling and analysis framework was proposed, in which regularisation and Random Forest …

Machine Learning As A Tool For Early Detection: A Focus On Late-Stage Colorectal Cancer Across Socioeconomic Spectrums, Hadiza Galadima, Rexford Anson-Dwamena, Ashley Johnson, Ghalib Bello, Georges Adunlin, James Blando

Community & Environmental Health Faculty Publications

Purpose: To assess the efficacy of various machine learning (ML) algorithms in predicting late-stage colorectal cancer (CRC) diagnoses against the backdrop of socio-economic and regional healthcare disparities. Methods: An innovative theoretical framework was developed to integrate individual- and census tract-level social determinants of health (SDOH) with sociodemographic factors. A comparative analysis of the ML models was conducted using key performance metrics such as AUC-ROC to evaluate their predictive accuracy. Spatio-temporal analysis was used to identify disparities in late-stage CRC diagnosis probabilities. Results: Gradient boosting emerged as the superior model, with the top predictors for late-stage CRC diagnosis being anatomic site, …

A Benchmark Framework For Data Visualization And Explainable Ai (Xai), Murat Kuzlu, Gokcen Ozdemir, Umut Ozdemir

Engineering Technology Faculty Publications

This research introduces a benchmark framework, called EDUMX, designed for machine learning (ML)- based forecasting and XAI tasks, leveraging the Streamlit open-source Python library. The framework offers a comprehensive suite of functionalities, including data loading, feature selection, relationship analysis, data preprocessing, model selection, metric evaluation, training, and real-time monitoring. Users can easily upload data in diverse formats, explore relationships between variables, preprocess data using various techniques, and assess the performance of the ML model using customizable metrics. With its user-friendly interface, this framework offers invaluable insights for forecasting tasks in various domains, catering to the evolving needs of predictive analytics. …

Nonuniform Sampling-Based Breast Cancer Classification, Santiago Posso

Theses and Dissertations--Electrical and Computer Engineering

The emergence of deep learning models and their success in visual object recognition have fueled the medical imaging community's interest in integrating these algorithms to improve medical diagnosis. However, natural images, which have been the main focus of deep learning models and mammograms, exhibit fundamental differences. First, breast tissue abnormalities are often smaller than salient objects in natural images. Second, breast images have significantly higher resolutions but are generally heavily downsampled to fit these images to deep learning models. Models that handle high-resolution mammograms require many exams and complex architectures. Additionally, spatially resizing mammograms leads to losing discriminative details essential …

Cross-Layer Design Of Highly Scalable And Energy-Efficient Ai Accelerator Systems Using Photonic Integrated Circuits, Sairam Sri Vatsavai

Theses and Dissertations--Electrical and Computer Engineering

Artificial Intelligence (AI) has experienced remarkable success in recent years, solving complex computational problems across various domains, including computer vision, natural language processing, and pattern recognition. Much of this success can be attributed to the advancements in deep learning algorithms and models, particularly Artificial Neural Networks (ANNs). In recent times, deep ANNs have achieved unprecedented levels of accuracy, surpassing human capabilities in some cases. However, these deep ANN models come at a significant computational cost, with billions to trillions of parameters. Recent trends indicate that the number of parameters per ANN model will continue to grow exponentially in the foreseeable …

Reinforcement Learning: Applying Low Discrepancy Action Selection To Deep Deterministic Policy Gradient, Aleksandr Svishchev

Electronic Theses and Dissertations

Reinforcement learning (RL) is a subfield of machine learning concerned with agents learning to behave optimally by interacting with an environment. One of the most important topics in RL is how the agent should explore, that is, how to choose actions in order to rate their impact on long-term reward. For example, a simple baseline strategy might be uniformly random action selection. This thesis investigates the heuristic idea that agents will learn faster if they explore by factoring the environment’s state into their decision and intentionally choose actions which are as different as possible from what they have previously observed. …

Locating Liability For Medical Ai, W. Nicholson Price Ii, I. Glenn Cohen

Articles

When medical AI systems fail, who should be responsible, and how? We argue that various features of medical AI complicate the application of existing tort doctrines and render them ineffective at creating incentives for the safe and effective use of medical AI. In addition to complexity and opacity, the problem of contextual bias, where medical AI systems vary substantially in performance from place to place, hampers traditional doctrines. We suggest instead the application of enterprise liability to hospitals—making them broadly liable for negligent injuries occurring within the hospital system—with an important caveat: hospitals must have access to the information needed …

Advancing Deep Learning With Graph-Based Structural Insights: From Graph Classification To Semantic Segmentation, Xin Ma

Computer Science and Engineering Dissertations

Deep learning has profoundly transformed machine learning by offering sophisticated data representations, yet effectively incorporating structural information remains a challenge. Structural data, whether explicit or implicit, has the potential to significantly enhance the performance of deep learning tasks. This research investigates the benefits of structural information across three crucial tasks: classification, clustering, and segmentation. For explicit structural data, where inputs are directly represented as graphs, we investigate graph-level classification in brain connectivity networks. We introduce the Multi-resolution Edge Network (MENET), a novel framework designed to identify disease-specific connectomic benchmarks with high discriminatory power across diagnostic categories. MENET leverages graph-level representations …

Music Recommendation Using Exemplars And Contrastive Learning, Tina Tran

Honors Undergraduate Theses

The popularity of AI audio applications is growing, it is used in chatbots, automated voice translation, virtual assistants, and text-to-speech translation. Audio classification is crucial in today’s world with a growing need to sort and classify millions of existing audio data with increasing amounts of new data uploaded over time. In the area of classification lies the difficult and lucrative problem of music recommendation. Research in music recommendation has trended over time towards collaborative-based approaches utilizing large amounts of user data. These approaches tend to deal with the cold-start problem of insufficient data and are costly to train. We look …

Accelerating Markov Chain Monte Carlo Sampling With Diffusion Models, N. T. Hunt-Smith, W. Melnitchouk, F. Ringer, N. Sato, A. W. Thomas, M. J. White

Physics Faculty Publications

Global fits of physics models require efficient methods for exploring high-dimensional and/or multimodal posterior functions. We introduce a novel method for accelerating Markov Chain Monte Carlo (MCMC) sampling by pairing a Metropolis-Hastings algorithm with a diffusion model that can draw global samples with the aim of approximating the posterior. We briefly review diffusion models in the context of image synthesis before providing a streamlined diffusion model tailored towards low-dimensional data arrays. We then present our adapted Metropolis-Hastings algorithm which combines local proposals with global proposals taken from a diffusion model that is regularly trained on the samples produced during the …

Diffusion Model Approach To Simulating Electron-Proton Scattering Events, Peter Devlin, Jian-Wei Qiu, Felix Ringer, Nobuo Sato

Physics Faculty Publications

Generative artificial intelligence is a fast-growing area of research offering various avenues for exploration in high-energy nuclear physics. In this work, we explore the use of generative models for simulating electron-proton collisions relevant to experiments like the Continuous Electron Beam Accelerator Facility and the future Electron-Ion Collider (EIC). These experiments play a critical role in advancing our understanding of nucleons and nuclei in terms of quark and gluon degrees of freedom. The use of generative models for simulating collider events faces several challenges such as the sparsity of the data, the presence of global or eventwide constraints, and steeply falling …

Flexible Attenuation Fields: Tomographic Reconstruction From Heterogeneous Datasets, Clifford S. Parker

Theses and Dissertations--Computer Science

Traditional reconstruction methods for X-ray computed tomography (CT) are highly constrained in the variety of input datasets they admit. Many of the imaging settings -- the incident energy, field-of-view, effective resolution -- remain fixed across projection images, and the only real variance is in the detector's position and orientation with respect to the scene. In contrast, methods for 3D reconstruction of natural scenes are extremely flexible to the geometric and photometric properties of the input datasets, readily accepting and benefiting from images captured under varying lighting conditions, with different cameras, and at disparate points in time and space. Extending CT …

Fr1: Comics, Cyborgs, And “In Between” Identities, Ella Lehavi

Scripps Senior Theses

As a queer Jew who grew up surrounded by immigrant cultures and communities, I find myself in a liminal space between my identities and the dominant culture of my country– one where my perspective on gender and my cultural experiences aren’t fully understood by the world I exist in. Comics and cartoons are an explorational platform for concepts of reality and identity; they are one of very few spaces where I see my identities explored with so much depth and care.

Cartoons and comics exist in between realistic depictions and abstraction. This makes them a great place to express all …

Statistically Principled Deep Learning For Sar Image Segmentation, Cassandra Goldberg

Honors Projects

This project explores novel approaches for Synthetic Aperture Radar (SAR) image segmentation that integrate established statistical properties of SAR into deep learning models. First, Perlin Noise and Generalized Gamma distribution sampling methods were utilized to generate a synthetic dataset that effectively captures the statistical attributes of SAR data. Subsequently, deep learning segmentation architectures were developed that utilize average pooling and 1x1 convolutions to perform statistical moment computations. Finally, supervised and unsupervised disparity-based losses were incorporated into model training. The experimental outcomes yielded promising results: the synthetic dataset effectively trained deep learning models for real SAR data segmentation, the statistically-informed architectures …

Adaptable And Trustworthy Machine Learning For Human Activity Recognition From Bioelectric Signals, Morgan S. Stuart

Theses and Dissertations

Enabling machines to learn measures of human activity from bioelectric signals has many applications in human-machine interaction and healthcare. However, labeled activity recognition datasets are costly to collect and highly varied, which challenges machine learning techniques that rely on large datasets. Furthermore, activity recognition in practice needs to account for user trust - models are motivated to enable interpretability, usability, and information privacy. The objective of this dissertation is to improve adaptability and trustworthiness of machine learning models for human activity recognition from bioelectric signals. We improve adaptability by developing pretraining techniques that initialize models for later specialization to unseen …

Low-Resource Automatic Speech Recognition Domain Adaptation – A Case-Study In Aviation Maintenance, Nadine Amin M.S., Tracy L. Yother Ph.D., Julia Rayz Ph.D.

Journal of Aviation/Aerospace Education & Research

With timeliness and efficiency being critical in the aviation maintenance industry, the need has been growing for smart technological solutions that optimize and streamline the different underlying tasks (Bergkvist & Sabbagh, 2021). One such task is the technical documentation of the performed maintenance operations (Chandola et al., 2022). Instead of manual documentation, voice tools that transcribe spoken logbook entries allow technicians to document their work right away in a hands-free and time efficient manner. However, an accurate automatic speech recognition (ASR) model requires large training corpora (Siyaev & Jo, 2021a), which are lacking in the domain of aviation maintenance. In …

Developing Policies For The Ethical Use Of Artificial Intelligence In Higher Education And Libraries, April Sheppard, Matthew Mayton

Staff and Faculty Scholarship

This presentation will provide sample artificial intelligence policy language from various higher education institutions and academic libraries. Topics covered will include the acceptable use of AI in the classroom, the role of faculty in making AI-related decisions, syllabus statements, AI use and detection, AI literacy, and library policies regarding AI. Participants will be able to compare and contrast policies to help them develop their own policies that work for their unique organization.

A Comparison Of Machine Learning Surrogate Models Of Street-Scale Flooding In Norfolk, Virginia, Diana Mcspadden, Steven Goldenberg, Binata Roy, Malachi Schram, Jonathan L. Goodall, Heather Richter

Community & Environmental Health Faculty Publications

Low-lying coastal cities, exemplified by Norfolk, Virginia, face the challenge of street flooding caused by rainfall and tides, which strain transportation and sewer systems and can lead to personal and property damage. While high-fidelity, physics-based simulations provide accurate predictions of urban pluvial flooding, their computational complexity renders them unsuitable for real-time applications. Using data from Norfolk rainfall events between 2016 and 2018, this study compares the performance of a previous surrogate model based on a random forest algorithm with two deep learning models: Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). The comparison of deep learning to the random …

A Survey On Few-Shot Class-Incremental Learning, Songsong Tian, Lusi Li, Weijun Li, Hang Ran, Xin Ning, Prayag Tiwari

Computer Science Faculty Publications

Large deep learning models are impressive, but they struggle when real-time data is not available. Few-shot class-incremental learning (FSCIL) poses a significant challenge for deep neural networks to learn new tasks from just a few labeled samples without forgetting the previously learned ones. This setup can easily leads to catastrophic forgetting and overfitting problems, severely affecting model performance. Studying FSCIL helps overcome deep learning model limitations on data volume and acquisition time, while improving practicality and adaptability of machine learning models. This paper provides a comprehensive survey on FSCIL. Unlike previous surveys, we aim to synthesize few-shot learning and incremental …

Charged Track Reconstruction With Artificial Intelligence For Clas12, Gagik Gavalian, Polykarpos Thomadakis, Angelos Angelopoulos, Nikos Chrisochoides

Computer Science Faculty Publications

In this paper, we present the results of charged particle track reconstruction in CLAS12 using artificial intelligence. In our approach, we use neural networks working together to identify tracks based on the raw signals in the Drift Chambers. A Convolutional Auto-Encoder is used to de-noise raw data by removing the hits that do not satisfy the patterns for tracks, and second Multi-Layer Perceptron is used to identify tracks from combinations of clusters in the drift chambers. Our method increases the tracking efficiency by 50% for multi-particle final states already conducted experiments. The de-noising results indicate that future experiments can run …

Short: Can Citations Tell Us About A Paper's Reproducibility? A Case Study Of Machine Learning Papers, Rochana R. Obadage, Sarah M. Rajtmajer, Jian Wu

Computer Science Faculty Publications

The iterative character of work in machine learning (ML) and artificial intelligence (AI) and reliance on comparisons against benchmark datasets emphasize the importance of reproducibility in that literature. Yet, resource constraints and inadequate documentation can make running replications particularly challenging. Our work explores the potential of using downstream citation contexts as a signal of reproducibility. We introduce a sentiment analysis framework applied to citation contexts from papers involved in Machine Learning Reproducibility Challenges in order to interpret the positive or negative outcomes of reproduction attempts. Our contributions include training classifiers for reproducibility-related contexts and sentiment analysis, and exploring correlations between …

Sub-Band Backdoor Attack In Remote Sensing Imagery, Kazi Aminul Islam, Hongyi Wu, Chunsheng Xin, Rui Ning, Liuwan Zhu, Jiang Li

Electrical & Computer Engineering Faculty Publications

Remote sensing datasets usually have a wide range of spatial and spectral resolutions. They provide unique advantages in surveillance systems, and many government organizations use remote sensing multispectral imagery to monitor security-critical infrastructures or targets. Artificial Intelligence (AI) has advanced rapidly in recent years and has been widely applied to remote image analysis, achieving state-of-the-art (SOTA) performance. However, AI models are vulnerable and can be easily deceived or poisoned. A malicious user may poison an AI model by creating a stealthy backdoor. A backdoored AI model performs well on clean data but behaves abnormally when a planted trigger appears in …