Computer Sciences Commons^™

Utilizing Few-Shot Meta Learning Algorithms For Medical Image Segmentation, Nick Littlefield

Thinking Matters Symposium

Deep learning models can be difficult to train because they require large amounts of data, which we usually do not have or are too expensive to get or annotate. To overcome this problem, we can use few-shot meta-learning, which allows us to train deep learning models with little data. Using a few examples, meta-learning, or learning-to-learn, aims to use the experience learned during training to generalize to unknown tasks. Medical imaging is an industry where it is particularly useful, as there is limited publicly available data due to patient privacy concerns and annotating costs.

This project examines how meta-learning performs …

Towards An Experimental Bibliography Of Hemispheric Reconstruction Newspapers, Joshua Ortiz Baco, Benjamin Charles Germain Lee, Jim Casey, Sarah H. Salter

Criticism

Digital collections of newspapers have drawn broader attention to the fragmented and scattered print histories of minoritized communities. Attempts to survey these histories through bibliography, however, quickly meet with a fundamental problem: the practice of bibliographic description calls for creating a static record of social affiliations. Given the overwhelming scholarly consensus that categories such as race, ethnicity, and language are socially constructed, this article introduces an experimental bibliographic method for mapping the vast landscape of historical newspapers. This method extends the machine learning affordances of a recent project called Newspaper Navigator to enumerate the newspapers in Chronicling America according to …

Stereotypes And Language Models: Understanding How Language Models Encode Stereotypes, Debiasing Language Models, And Examining How Stereotypes Affect Conversations, Brian C. Wang

Computer Science Senior Theses

This thesis describes a variety of approaches in examining how language models encode stereotypes (understanding stereotypes from a model point-of-view), debiasing language models, and using language models to understand how stereotypes affect conversations (understanding stereotypes from a conversational point-of-view). We present a novel approach for textual clues analysis that makes language models more interpretable, combining the understanding of what stereotypes the internal structures of language models have encoded during their initial training (via attention-based analysis) and understanding what textual clues are most relevant to identifying stereotypes for models trained to detect stereotypes (via SHAP-based analysis). We find that different pre-trained …

Exploiting Association Rules Mining To Inform The Use Of Non-Manual Features In Sign Language Processing, Robert G. Smith

Other Resources

In recent years, the use of virtual assistants and voice user interfaces has become a latent part of modern living. Unseen to the user are the various artificial intelligence and natural language processing technologies, the vast datasets, and the linguistic insights that underpin such tools. The technologies supporting them have chiefly targeted widely used spoken languages, leaving sign language users at a disadvantage. One important reason why sign languages are unsupported by such tools is a requirement of the underpinning technologies for a comprehensive description of the language. Sign language processing technologies endeavour to bridge this technology inequality.

Recent approaches …

Improving The Efficiency Of Exponential Ratio-Type Estimator For Population Median: A Calibration Weight Adjustment Approach, Mathew J. Iseh, Kufre J. Bassey

CBN Journal of Applied Statistics (JAS)

This paper modifies the Bahl and Tuteja exponential ratio-type estimator for population median under simple random and stratified sampling schemes using calibration weight adjustment technique with supplementary information to vary the stratum weights. The bias and mean square error of the modified estimator were obtained up to the second-order approximation, which satisfies the necessary conditions for efficiency. The findings show that the new estimator surpasses existing estimators in efficiency gain. This suggests the appropriateness of calibration weight modification in boosting the efficiency of a population parameter estimator under stratified random sampling especially where the population parameter of the auxiliary variable …

Mosaic: Spatially-Multiplexed Edge Ai Optimization Over Multiple Concurrent Video Sensing Streams, Ila Gokarn, Hemanth Sabbella, Yigong Hu, Tarek Abdelzaher, Archan Misra

Research Collection School Of Computing and Information Systems

Sustaining high fidelity and high throughput of perception tasks over vision sensor streams on edge devices remains a formidable challenge, especially given the continuing increase in image sizes (e.g., generated by 4K cameras) and complexity of DNN models. One promising approach involves criticality-aware processing, where the computation is directed selectively to "critical" portions of individual image frames. We introduce MOSAIC, a novel system for such criticality-aware concurrent processing of multiple vision sensing streams that provides a multiplicative increase in the achievable throughput with negligible loss in perception fidelity. MOSAIC determines critical regions from images received from multiple vision …

Trustworthy Machine Learning Through The Lens Of Privacy And Security, Thi Kim Phung Lai

Dissertations

Nowadays, machine learning (ML) becomes ubiquitous and it is transforming society. However, there are still many incidents caused by ML-based systems when ML is deployed in real-world scenarios. Therefore, to allow wide adoption of ML in the real world, especially in critical applications such as healthcare, finance, etc., it is crucial to develop ML models that are not only accurate but also trustworthy (e.g., explainable, privacy-preserving, secure, and robust). Achieving trustworthy ML with different machine learning paradigms (e.g., deep learning, centralized learning, federated learning, etc.), and application domains (e.g., computer vision, natural language, human study, malware systems, etc.) is challenging, …

Deep Hybrid Modeling Of Neuronal Dynamics Using Generative Adversarial Networks, Soheil Saghafi

Dissertations

Mechanistic modeling and machine learning methods are powerful techniques for approximating biological systems and making accurate predictions from data. However, when used in isolation these approaches suffer from distinct shortcomings: model and parameter uncertainty limit mechanistic modeling, whereas machine learning methods disregard the underlying biophysical mechanisms. This dissertation constructs Deep Hybrid Models that address these shortcomings by combining deep learning with mechanistic modeling. In particular, this dissertation uses Generative Adversarial Networks (GANs) to provide an inverse mapping of data to mechanistic models and identifies the distributions of mechanistic model parameters coherent to the data.

Chapter 1 provides background information on …

Data-Optimized Spatial Field Predictions For Robotic Adaptive Sampling: A Gaussian Process Approach, Zachary Nathan

Computer Science Senior Theses

We introduce a framework that combines Gaussian Process models, robotic sensor measurements, and sampling data to predict spatial fields. In this context, a spatial field refers to the distribution of a variable throughout a specific area, such as temperature or pH variations over the surface of a lake. Whereas existing methods tend to analyze only the particular field(s) of interest, our approach optimizes predictions through the effective use of all available data. We validated our framework on several datasets, showing that errors can decline by up to two-thirds through the inclusion of additional colocated measurements. In support of adaptive sampling, …

A Survey On Online Matching And Ad Allocation, Ryan Lee

Theses

One of the classical problems in graph theory is matching. Given an undirected graph, find a matching which is a set of edges without common vertices. In 1990s, Richard Karp, Umesh Vazirani, and Vijay Vazirani would be the first computer scientists to use matchings for online algorithms [8]. In our domain, an online algorithm operates in the online setting where a bipartite graph is given. On one side of the graph there is a set of advertisers and on the other side we have a set of impressions. During the online phase, multiple impressions will arrive and the objective of …

"Church On My Couch": Predicting The Future Impact Of Online Ministry Based On The Impact During Covid-19, Samukeliso Mabarani, Sikhumbuzo Dube

Adventist Human-Subject Researchers Association

With “everything from home” as the new norm, “how does the use of digital platforms impact Adventist education, community engagement, and spiritual outreach?” Using a quantitative approach, we draw insights from online ministry during Covid-19 and use the insights to predict the future impact of online ministry statistically.

Covid-19 In Casinos: Analysis Of Covid-19 Contamination And Spread With Economic Impact Assessment, Anastasia (Stasi) D. Baran, Jason D. Fiege

International Conference on Gambling & Risk Taking

Abstract:

The COVID-19 pandemic caused tremendous disruption for casinos, with the virus causing various lengths of shutdowns, capacity restrictions, and social distancing strategies such as machine removals or section closures. Although most of the world has now eased off these measures, it is important to review lessons learned to understand, and better prepare for similar circumstances in the future. We present Monte Carlo slot floor simulation software customized to simulate players spreading COVID-19 on the slot floor. We simulate the amount of touch surface contamination; the number of potential surface contact exposure events per day, and a proximity exposures statistic …

Statistical Methods To Generate Artificial Slot Floor Data For The Advancement Of Casino Related Research, Courtney Bonner, Anastasia (Stasi) D. Baran, Jason D. Fiege, Saman Muthukumarana

International Conference on Gambling & Risk Taking

Abstract:

A common difficulty when researching gambling topics is the availability of high-quality data sets for development and testing. Due to the high level of secrecy within the gambling industry, if data is obtained for research purposes it is often prohibitively obfuscated, incomplete, or aggregated. Although these data have allowed for advancement in academic work, it leaves both the researchers and readers left wondering about what would be possible if more detailed data sets were available. To mitigate the paucity of data available to researchers, we present a Markov chain-based statistical process for producing artificial event data for a simulated …

The Locals Casino As A Social Network – Can An Interconnected Community Of Players Detect Differences In Hold?, Jason D. Fiege, Anastasia (Stasi) D. Baran

International Conference on Gambling & Risk Taking

Abstract

It is difficult for individual players to detect differences in theoretical hold between slot machines without playing an unrealistically large number of games. This difficulty occurs because the fractional loss incurred by a player converges only slowly to the theoretical hold in the presence of volatility designed into slot pay tables. Nevertheless, many operators believe that players can detect changes in hold or differences compared to competition, especially in a locals casino market, and therefore resist increasing holds. Instead of investigating whether individual players can detect differences in hold, we ask whether a population of casino regulars who share …

Algorithmic Bias: Causes And Effects On Marginalized Communities, Katrina M. Baha

Undergraduate Honors Theses

Individuals from marginalized backgrounds face different healthcare outcomes due to algorithmic bias in the technological healthcare industry. Algorithmic biases, which are the biases that arise from the set of steps used to solve or analyze a problem, are evident when people from marginalized communities use healthcare technology. For example, many pulse oximeters, which are the medical devices used to measure oxygen saturation in the blood, are not able to accurately read people who have darker skin tones. Thus, people with darker skin tones are not able to receive proper health care due to their pulse oximetry data being inaccurate. This …

Automated Delineation Of Visual Area Boundaries And Eccentricities By A Cnn Using Functional, Anatomical, And Diffusion-Weighted Mri Data, Noah C. Benson, Bogeng Song, Toshikazu Miyata, Hiromasa Takemura, Jonathan Winawer

MODVIS Workshop

Delineating visual field maps and iso-eccentricities from fMRI data is an important but time-consuming task for many neuroimaging studies on the human visual cortex because the traditional methods of doing so using retinotopic mapping experiments require substantial expertise as well as scanner, computer, and human time. Automated methods based on gray-matter anatomy or a combination of anatomy and functional mapping can reduce these requirements but are less accurate than experts. Convolutional Neural Networks (CNNs) are powerful tools for automated medical image segmentation. We hypothesize that CNNs can define visual area boundaries with high accuracy. We trained U-Net CNNs with ResNet18 …

Ptmtorrent: A Dataset For Mining Open-Source Pre-Trained Model Packages, Wenxin Jiang, Nicholas Synovic, Purvish Jajal, Taylor R. Schorlemmer, Arav Tewari, Bhavesh Pareek, George K. Thiruvathukal, James C. Davis

Computer Science: Faculty Publications and Other Works

Due to the cost of developing and training deep learning models from scratch, machine learning engineers have begun to reuse pre-trained models (PTMs) and fine-tune them for downstream tasks. PTM registries known as “model hubs” support engineers in distributing and reusing deep learning models. PTM packages include pre-trained weights, documentation, model architectures, datasets, and metadata. Mining the information in PTM packages will enable the discovery of engineering phenomena and tools to support software engineers. However, accessing this information is difficult — there are many PTM registries, and both the registries and the individual packages may have rate limiting for accessing …

Pruning Ghsom To Create An Explainable Intrusion Detection System, Thomas Michael Kirby

Theses and Dissertations

Intrusion Detection Systems (IDS) that provide high detection rates but are black boxes lead
to models that make predictions a security analyst cannot understand. Self-Organizing Maps
(SOMs) have been used to predict intrusion to a network, while also explaining predictions through
visualization and identifying significant features. However, they have not been able to compete with
the detection rates of black box models. Growing Hierarchical Self-Organizing Maps (GHSOMs)
have been used to obtain high detection rates on the NSL-KDD and CIC-IDS-2017 network traffic
datasets, but they neglect creating explanations or visualizations, which results in another black
box model.
This paper offers …

An Analysis Of Text-Based Machine Learning Models For Vulnerability Detection, Kollin Ryne Napier

Theses and Dissertations

With an increase in complexity of software, developers rely more on reuse and dependencies in their source code via code snippets. As a result, it is becoming harder to identify and mitigate vulnerabilities. Although traditional analysis tools are still utilized, machine learning models are being adopted to expand efforts and combat such threats. Given the possibilities towards usage of such models, research in this area has introduced various approaches which vary in usability and prediction. In generalizing models to a more natural language approach, researchers have opted to train models on source code to identify existing and potential vulnerabilities. Exploratory …

Big Ideas In Sports Analytics And Statistical Tools For Their Investigation, Benjamin S. Baumer, Gregory J. Matthews, Quang Nguyen

Statistical and Data Sciences: Faculty Publications

Sports analytics—broadly defined as the pursuit of improvement in athletic performance through the analysis of data—has expanded its footprint both in the professional sports industry and in academia over the past 30 years. In this article, we connect four big ideas that are common across multiple sports: the expected value of a game state, win probability, measures of team strength, and the use of sports betting market data. For each, we explore both the shared similarities and individual idiosyncracies of analytical approaches in each sport. While our focus is on the concepts underlying each type of analysis, any implementation necessarily …

Brif: A Novel And Efficient Implementation Of Random Forests Based On Bit Packing And Parallel Computing, Yanchao Liu

Industrial and Systems Engineering Faculty Research Publications

Random forests are powerful and popular machine learning methods. While general principles of tree induction are straightforward and well-understood, the numerous algorithmic treatments implemented in software tools, as well as their impacts on performance, are less familiar to most users. This paper introduces a new random forest toolkit (the ‘brif’ package in R and Python) along with its key algorithmic design features, and demonstrates the effects of the forest’s hyper-parameters such as the split search method, tree depth and the voting mechanism, on the classification performance. Summaries of benchmarking experiments are also presented. Results show that ‘brif’ stands out among …

Machine Learning-Based Data And Model Driven Bayesian Uncertanity Quantification Of Inverse Problems For Suspended Non-Structural System, Zhiyuan Qin

All Dissertations

Inverse problems involve extracting the internal structure of a physical system from noisy measurement data. In many fields, the Bayesian inference is used to address the ill-conditioned nature of the inverse problem by incorporating prior information through an initial distribution. In the nonparametric Bayesian framework, surrogate models such as Gaussian Processes or Deep Neural Networks are used as flexible and effective probabilistic modeling tools to overcome the high-dimensional curse and reduce computational costs. In practical systems and computer models, uncertainties can be addressed through parameter calibration, sensitivity analysis, and uncertainty quantification, leading to improved reliability and robustness of decision and …

A Programmatic Geographic Information Systems Analysis Of Plant Hardiness Zones, Andrew Bowen

Electronic Theses and Dissertations

The Plant Hardiness Zone Map consists of thirteen geographical zones that describe whether a plant can survive based on average annual minimal temperatures. As climate change progresses, minimum temperatures in all regions are expected to change. This work programmatically evaluates predicted future climate projection data and converts it to United States Department of Agriculture-defined hardiness zones. Through the next 80 years, hardiness zones are projected to move poleward; in effect, colder zones will lose area and warmer zones will gain area globally. Some implications include changes in crop growing degree days, which could alter crop productivity, migration and settlement of …

Predicting High-Cap Tech Stock Polarity: A Combined Approach Using Support Vector Machines And Bidirectional Encoders From Transformers, Ian L. Grisham

Electronic Theses and Dissertations

The abundance, accessibility, and scale of data have engendered an era where machine learning can quickly and accurately solve complex problems, identify complicated patterns, and uncover intricate trends. One research area where many have applied these techniques is the stock market. Yet, financial domains are influenced by many factors and are notoriously difficult to predict due to their volatile and multivariate behavior. However, the literature indicates that public sentiment data may exhibit significant predictive qualities and improve a model’s ability to predict intricate trends. In this study, momentum SVM classification accuracy was compared between datasets that did and did not …

A Study Of Various Data Sizes Using Machine Learning, Sochaeta Koeum

Electronic Theses, Projects, and Dissertations

Social media is a great domain for news consumption; however, it is referred to as a double-edged sword. While it is user-friendly and low-cost, social media is the reason why fake news can spread rapidly, which is detrimental to society, businesses, and many consumers. Therefore, fake news detection is an emerging field. However, some challenges have restricted other researchers from developing a universal machine learning model that is fast, efficient, and reliable to stop the proliferation because of the lack of resources available, such as large-sized datasets. The goal of this culminating experience project is to explore how varying datasets …

Heart Disease Prediction Using Binary Classification, Virendra Sunil Devare

Electronic Theses, Projects, and Dissertations

In this project, I built a neural network model to predict heard disease with binary classification technique using patient information dataset from UCI Machine Learning repository. This dataset was preprocessed to remove missing elements and performed feature extraction. Our result shows that the model that I built has the best performance accuracy in heart disease classification if compared to other models and algorithms. The model achieved 94.98% accuracy after hyperparameter tuning and 0.947 area under the curve in ROC curve analysis. In addition, to identify the most important factors in heart disease prediction, I also performed feature importance analysis. Our …

Multiparametric Magnetic Resonance Imaging Artificial Intelligence Pipeline For Oropharyngeal Cancer Radiotherapy Treatment Guidance, Kareem Wahid

Dissertations and Theses (Open Access)

Oropharyngeal cancer (OPC) is a widespread disease and one of the few domestic cancers that is rising in incidence. Radiographic images are crucial for assessment of OPC and aid in radiotherapy (RT) treatment. However, RT planning with conventional imaging approaches requires operator-dependent tumor segmentation, which is the primary source of treatment error. Further, OPC expresses differential tumor/node mid-RT response (rapid response) rates, resulting in significant differences between planned and delivered RT dose. Finally, clinical outcomes for OPC patients can also be variable, which warrants the investigation of prognostic models. Multiparametric MRI (mpMRI) techniques that incorporate simultaneous anatomical and functional information …

Automating The Radiation Therapy Treatment Planning Process For Pediatric Patients With Medulloblastoma, Soleil Hernandez

Dissertations and Theses (Open Access)

Over the past 50 years, pediatric cancer 5-year survival rates increased from 20% to 80% in high-income countries, however, these trends have not been mirrored in low-and-middle-income countries (LMICs). This is due in part to delayed diagnosis, higher rates of advanced disease at presentation and a growing lack of access to high quality medical personnel and technology necessary to deliver complex treatments.

The long-term goal of this study was to alleviate demanding workflows and increase global access to high-quality pediatric radiation therapy by harnessing the power of artificial intelligence to automate the radiation therapy treatment planning process for pediatric patients …

Liloc: Enabling Precise 3d Localization In Dynamic Indoor Environments Using Lidars, Darshana Rathnayake, Meera Radhakrishnan, Inseok Hwang, Archan Misra

Research Collection School Of Computing and Information Systems

We present LiLoc, a system for precise 3D localization and tracking of mobile IoT devices (e.g., robots) in indoor environments using multi-perspective LiDAR sensing. The key differentiators in our work are: (a) First, unlike traditional localization approaches, our approach is robust to dynamically changing environmental conditions (e.g., varying crowd levels, object placement/layout changes); (b) Second, unlike prior work on visual and 3D SLAM, LiLoc is not dependent on a pre-built static map of the environment and instead works by utilizing dynamically updated point clouds captured from both infrastructural-mounted LiDARs and LiDARs equipped on individual mobile IoT devices. To achieve fine-grained, …

Quantification Of Various Types Of Biases In Large Language Models, Sudhashree Sayenju

Doctor of Data Science and Analytics Dissertations

Natural Language Processing (NLP) systems are included everywhere on the internet from search engines, language translations to more advanced systems like voice assistant and customer service. Since humans are always on the receiving end of NLP technologies, it is very important to analyze whether or not the Large Language Models (LLMs) in use have bias and are therefore unfair. The majority of the research in NLP bias has focused on societal stereotype biases embedded in LLMs. However, our research focuses on all types of biases, namely model class level bias, stereotype bias and domain bias present in LLMs. Model class …