Physical Sciences and Mathematics Commons^™

Information-Theoretic Model Diagnostics (Infomod), Armin Esmaeilzadeh

UNLV Theses, Dissertations, Professional Papers, and Capstones

Model validation is a critical step in the development, deployment, and governance of machine learning models. During the validation process, the predictive power of a model is measured on unseen datasets with a variety of metrics such as Accuracy and F1-Scores for classification tasks. Although the most used metrics are easy to implement and understand, they are aggregate measures over all the segments of heterogeneous datasets, and therefore, they do not identify the performance variation of a model among different data segments. The lack of insight into how the model performs over segments of unseen datasets has raised significant challenges …

Transient Sources And How To Study Them: Selected Topics In Multi-Messenger Astronomy, Jiawei Luo

UNLV Theses, Dissertations, Professional Papers, and Capstones

The discovery of cosmic neutrino flux by IceCube, and the multi-messenger observations of gravitational event GW170817 ushered in the era of multi-messenger astronomy. Since the Universe itself is a natural laboratory, multi-messenger astronomy can help us study the most extreme physics processes in great detail. In this dissertation, we touch on some of the currently unanswered questions involving different types of transient sources and different “messengers” of multi-messenger astronomy. We employ a variety of analysis methods, including machine learning, a method that has not yet been widely adopted in astronomy but is rapidly gaining momentum.We start this dissertation with Chapter …

An Introduction To Federated Learning And Its Analysis, Manjari Ganapathy

UNLV Theses, Dissertations, Professional Papers, and Capstones

With the onset of the digital era, data privacy is one of the most predominant issues. Decentralized learning is becoming popular as the data can remain within local entities by maintaining privacy. Federated Learning is a decentralized machine learning approach, where multiple clients collaboratively learn a model, without sharing raw data. There are many practical challenges in solving Federated Learning, which include communication set up, data heterogeneity and computational capacity of clients. In this thesis, I explore recent methods of Federated Learning with various settings, such as data distributions and data variability, used in several applications. In addition, I, specifically, …

A Framework For Vector-Weighted Deep Neural Networks, Carter Chiu

UNLV Theses, Dissertations, Professional Papers, and Capstones

The vast majority of advances in deep neural network research operate on the basis of a real-valued weight space. Recent work in alternative spaces have challenged and complemented this idea; for instance, the use of complex- or binary-valued weights have yielded promising and fascinating results. We propose a framework for a novel weight space consisting of vector values which we christen VectorNet. We first develop the theoretical foundations of our proposed approach, including formalizing the requisite theory for forward and backpropagating values in a vector-weighted layer. We also introduce the concept of expansion and aggregation functions for conversion between real …

Machine Learning Classification Of Primary Tissue Origin Of Cancer From Dna Methylation Markers, Sravani Gannavarapu Surya Naga

UNLV Theses, Dissertations, Professional Papers, and Capstones

Cancer is one of the leading causes of death globally and was responsible for approximately 9.6 million deaths in 2018. One of the main reason for deaths from cancer is late-stage presentation and inaccessible diagnosis and treatment. Cancer often spreads from the part of the body where it started (primary site) to a different part of the body (metastatic site). Identifying the primary site of cancer plays a key role as it directs the appropriate treatment. Cancer which spreads needs the same treatment as its origin. Having this knowledge can help doctors to decide the type of treatment.

All cancers …

Machine Learning Applications In Graduation Prediction At The University Of Nevada, Las Vegas, Elliott Collin Ploutz

UNLV Theses, Dissertations, Professional Papers, and Capstones

Graduation rates of four-year institutions are an increasingly important metric to incoming students and for ranking universities. To increase completion rates, universities must analyze available student data to understand trends and factors leading to graduation. Using predictive modeling, incoming students can be assessed as to their likelihood of completing a degree. If students are predicted to be most likely to drop out, interventions can be enacted to increase retention and completion rates.

At the University of Nevada, Las Vegas (UNLV), four-year graduation rates are 15% and six-year graduation rates are 39%. To improve these rates, we have gathered seven years …

A Test Driven Approach To Develop Web-Based Machine Learning Applications, Armin Esmaeilzadeh

UNLV Theses, Dissertations, Professional Papers, and Capstones

The purpose of this thesis is to propose the design and architecture of a testable, scalable, and ef-cient web-based application that models and implements machine learning applications in cancer prediction. There are various components that form the architecture of our web-based application including server, database, programming language, web framework, and front-end design. There are also other factors associated with our application such as testability, scalability, performance, and design pattern. Our main focus in this thesis is on the testability of the system while consid- ering the importance of other factors as well.

The data set for our application is a …

A Comparative Study On Text Categorization, Aditya Chainulu Karamcheti

UNLV Theses, Dissertations, Professional Papers, and Capstones

Automated text categorization is a supervised learning task, defined as assigning category labels to new documents based on likelihood suggested by a training set of labeled documents. Two examples of methodology for text categorizations are Naive Bayes and K-Nearest Neighbor.

In this thesis, we implement two categorization engines based on Naive Bayes and K-Nearest Neighbor methodology. We then compare the effectiveness of these two engines by calculating standard precision and recall for a collection of documents. We will further report on time efficiency of these two engines.