This work details the bootstrap estimation of a nonparametric information divergence measure, the Dp divergence measure, using a power law model. To address the challenge posed by computing accurate divergence estimates given finite size data, the bootstrap approach is used in conjunction with a power law curve to calculate an asymptotic value of the divergence estimator. Monte Carlo estimates of Dp are found for increasing values of sample size, and a power law fit is used to relate the divergence estimates as a function of sample size. The fit is also used to generate a confidence interval for the estimate to characterize the quality of the estimate. We compare the performance of this method with the other estimation methods. The calculated divergence is applied to the binary classification problem. Using the inherent relation between divergence measures and classification error rate, an analysis of the Bayes error rate of several data sets is conducted using the asymptotic divergence estimate.

Divergence functions are both highly useful and fundamental to many areas in information theory and machine learning, but require either parametric approaches or prior knowledge of labels on the full data set. This paper presents a method to estimate the divergence between two data sets in the absence of fully labeled data. This semi-labeled case is common in many domains where labeling data by hand is expensive or time-consuming, or wherever large data sets are present. The theory derived in this paper is demonstrated on a simulated example, and then applied to a feature selection and classification problem from pathological speech analysis.

Epilepsy affects numerous people around the world and is characterized by recurring seizures, prompting the ability to predict them so precautionary measures may be employed. One promising algorithm extracts spatiotemporal correlation based features from intracranial electroencephalography signals for use with support vector machines. The robustness of this methodology is tested through a sensitivity analysis. Doing so also provides insight about how to construct more effective feature vectors.

We present in this paper a method to compare scene classification accuracy of C-band Synthetic aperture radar (SAR) and optical images utilizing both classical and quantum computing algorithms. This REU study uses data from the Sentinel satellite. The dataset contains (i) synthetic aperture radar images collected from the Sentinel-1 satellite and (ii) optical images for the same area as the SAR images collected from the Sentinel-2 satellite. We utilize classical neural networks to classify four classes of images. We then use Quantum Convolutional Neural Networks and deep learning techniques to take advantage of machine learning to help the system train, learn, and identify at a higher classification accuracy. A hybrid Quantum-classical model that is trained on the Sentinel1-2 dataset is proposed, and the performance is then compared against the classical in terms of classification accuracy.

Classification in machine learning is quite crucial to solve many problems that the world is presented with today. Therefore, it is key to understand one’s problem and develop an efficient model to achieve a solution. One technique to achieve greater model selection and thus further ease in problem solving is estimation of the Bayes Error Rate. This paper provides the development and analysis of two methods used to estimate the Bayes Error Rate on a given set of data to evaluate performance. The first method takes a “global” approach, looking at the data as a whole, and the second is more “local”—partitioning the data at the outset and then building up to a Bayes Error Estimation of the whole. It is found that one of the methods provides an accurate estimation of the true Bayes Error Rate when the dataset is at high dimension, while the other method provides accurate estimation at large sample size. This second conclusion, in particular, can have significant ramifications on “big data” problems, as one would be able to clarify the distribution with an accurate estimation of the Bayes Error Rate by using this method.

This paper serves to report the research performed towards detecting PD and the effects of medication through the use of machine learning and finger tapping data collected through mobile devices. The primary objective for this research is to prototype a PD classification model and a medication classification model that predict the following: the individual’s disease status and the medication intake time relative to performing the finger-tapping activity, respectively.

This paper serves to report the research performed towards detecting PD and the effects of medication through the use of machine learning and finger tapping data collected through mobile devices. The primary objective for this research is to prototype a PD classification model and a medication classification model that predict the following: the individual’s disease status and the medication intake time relative to performing the finger-tapping activity, respectively.

In wireless communication systems, the process of data transmission includes the estimation of channels. Implementing machine learning in this process can reduce the amount of time it takes to estimate channels, thus, resulting in an increase of the system’s transmission throughput. This maximizes the performance of applications relating to device-to-device communications and 5G systems. However, applying machine learning algorithms to multi-base-station systems is not well understood in literature, which is the focus of this thesis.

The ability of magnetic resonance imaging (MRI) to image any part of the human body without the effects of harmful radiation such as in CAT and PET scans established MRI as a clinical mainstay for a variety of different ailments and maladies. Short wavelengths accompany the high frequencies present in high-field MRI, and are on the same scale as the human body at a static magnetic field strength of 3 T (128 MHz). As a result of these shorter wavelengths, standing wave effects are produced in the MR bore where the patient is located. These standing waves generate bright and dark spots in the resulting MR image, which correspond to irregular regions of high and low clarity. Coil loading is also an inevitable byproduct of subject positioning inside the bore, which decreases the signal that the region of interest (ROI) receives for the same input power. Several remedies have been proposed in the literature to remedy the standing wave effect, including the placement of high permittivity dielectric pads (HPDPs) near the ROI. Despite the success of HPDPs at smoothing out image brightness, these pads are traditionally bulky and take up a large spatial volume inside the already small MR bore. In recent years, artificial periodic structures known as metamaterials have been designed to exhibit specific electromagnetic effects when placed inside the bore. Although typically thinner than HPDPs, many metamaterials in the literature are rigid and cannot conform to the shape of the patient, and some are still too bulky for practical use in clinical settings. The well-known antenna engineering concept of fractalization, or the introduction of self-similar patterns, may be introduced to the metamaterial to display a specific resonance curve as well as increase the metamaterial’s intrinsic capacitance. Proposed in this paper is a flexible fractal-inspired metamaterial for application in 3 T MR head imaging. To demonstrate the advantages of this flexibility, two different metamaterial configurations are compared to determine which produces a higher localized signal-to-noise ratio (SNR) and average signal measured in the image: in the first configuration, the metamaterial is kept rigid underneath a human head phantom to represent metamaterials in the literature (single-sided placement); and in the second, the metamaterial is wrapped around the phantom to utilize its flexibility (double-sided placement). The double-sided metamaterial setup was found to produce an increase in normalized SNR of over 5% increase in five of six chosen ROIs when compared to no metamaterial use and showed a 10.14% increase in the total average signal compared to the single-sided configuration.