Theory and Algorithms Commons^™

Meta Learning Recommendation System For Classification, Clarence O. Williams Iii

Theses and Dissertations

A data driven approach is an emerging paradigm for the handling of analytic problems. In this paradigm the mantra is to let the data speak freely. However, when using machine learning algorithms, the data does not naturally reveal the best or even a good approach for algorithm choice. One method to let the algorithm reveal itself is through the use of Meta Learning, which uses the features of a dataset to determine a useful model to represent the entire dataset. This research proposes an improvement on the meta-model recommendation system by adding classification problems to the candidate problem space with …

Leveraging Peer-To-Peer Energy Sharing For Resource Optimization In Mobile Social Networks, Aashish Dhungana

Theses and Dissertations

Mobile Opportunistic Networks (MSNs) enable the interaction of mobile users in the vicinity through various short-range wireless communication technologies (e.g., Bluetooth, WiFi) and let them discover and exchange information directly or in ad hoc manner. Despite their promise to enable many exciting applications, limited battery capacity of mobile devices has become the biggest impediment to these appli- cations. The recent breakthroughs in the areas of wireless power transfer (WPT) and rechargeable lithium batteries promise the use of peer-to-peer (P2P) energy sharing (i.e., the transfer of energy from the battery of one member of the mobile network to the battery of …

Sparsity And Weak Supervision In Quantum Machine Learning, Seyran Saeedi

Theses and Dissertations

Quantum computing is an interdisciplinary field at the intersection of computer science, mathematics, and physics that studies information processing tasks on a quantum computer. A quantum computer is a device whose operations are governed by the laws of quantum mechanics. As building quantum computers is nearing the era of commercialization and quantum supremacy, it is essential to think of potential applications that we might benefit from. Among many applications of quantum computation, one of the emerging fields is quantum machine learning. We focus on predictive models for binary classification and variants of Support Vector Machines that we expect to be …

Invariance And Invertibility In Deep Neural Networks, Han Zhang

Theses and Dissertations

Machine learning is concerned with computer systems that learn from data instead of being explicitly programmed to solve a particular task. One of the main approaches behind recent advances in machine learning involves neural networks with a large number of layers, often referred to as deep learning. In this dissertation, we study how to equip deep neural networks with two useful properties: invariance and invertibility. The first part of our work is focused on constructing neural networks that are invariant to certain transformations in the input, that is, some outputs of the network stay the same even if the input …

Methodology For Comparison Of Algorithms For Real-World Multi-Objective Optimization Problems: Space Surveillance Network Design, Troy B. Dontigney

Theses and Dissertations

Space Situational Awareness (SSA) is an activity vital to protecting national and commercial satellites from damage or destruction due to collisions. Recent research has demonstrated a methodology using evolutionary algorithms (EAs) which is intended to develop near-optimal Space Surveillance Network (SSN) architectures in the sense of low cost, low latency, and high resolution. That research is extended here by (1) developing and applying a methodology to compare the performance of two or more algorithms against this problem, and (2) analyzing the effects of using reduced data sets in those searches. Computational experiments are presented in which the performance of five …

Two-On-One Pursuit With A Non-Zero Capture Radius, Patrick J. Wasz

Theses and Dissertations

In this paper, we revisit the "Two Cutters and Fugitive Ship" differential game that was addressed by Isaacs, but move away from point capture. We consider a two-on-one pursuit-evasion differential game with simple motion and pursuers endowed with circular capture sets of radius l > 0. The regions in the state space where only one pursuer effects the capture and the region in the state space where both pursuers cooperatively and isochronously capture the evader are characterized, thus solving the Game of Kind. Concerning the Game of Degree, the algorithm for the synthesis of the optimal state feedback strategies of the …

A Generalized Phase Gradient Autofocus Algorithm, Aaron Evers

Theses and Dissertations

The phase gradient autofocus (PGA) algorithm has seen widespread use and success within the synthetic aperture radar (SAR) imaging community. However, its use and success has largely been limited to collection geometries where either the polar format algorithm (PFA) or range migration algorithm is suitable for SAR image formation. In this work, a generalized phase gradient autofocus (GPGA) algorithm is developed which is applicable with both the PFA and backprojection algorithm (BPA), thereby directly supporting a wide range of collection geometries and SAR imaging modalities. The GPGA algorithm preserves the four crucial signal processing steps comprising the PGA algorithm, while …

Performance Analysis Of Angle Of Arrival Algorithms Applied To Radiofrequency Interference Direction Finding, Taylor S. Barber

Theses and Dissertations

Radiofrequency (RF) interference threatens the functionality of systems that increasingly underpin the daily function of modern society. In recent years there have been multiple incidents of intentional RF spectrum denial using terrestrial interference sources. Because RF based systems are used in safety-of-life applications in both military and civilian contexts, there is need for systems that can quickly locate these interference sources. In order to meet this need, the Air Force Research Laboratory Weapons Directorate is sponsoring the following research to support systems that will be able to quickly geolocate RF interferers using passive angle-of-arrival estimation to triangulate interference sources. This …

The Effect Of Modeling Simultaneous Events On Simulation Results, John M. Carboni

Theses and Dissertations

This thesis explores the method that governs the prioritizing process for simultaneous events in relation to simulation results for discrete-event simulations. Specifically, it contrasts typical discrete-event simulation (DES) execution algorithms with how events are selected and ordered by the discrete-event system specification (DEVS) formalism. The motivation for this research stems from a desire to understand how the selection of events affects simulation output (i.e., response). As a particular use case, we briefly investigate the processing of simultaneous events by the Advanced Framework for Simulation, Integration and Modeling (AFSIM), a military discrete-event combat modeling and simulation package. To facilitate the building …

A Simulation Model For Estimating Human Error Probability, Nitisha Reddy Boyapati

Theses and Dissertations

This report describes the system dynamics architecture of a simulation model which estimates human error probability for humans performing certain tasks in a given scenario. Human error probability is estimated as a function of the type of tasks performed and the number of performance shaping factors. In this work, the Standardized Plant Analysis Risk-Human (SPAR-H) reliability analysis method is utilized for estimating the probability of human error. The system dynamics simulation model captures the cause and effect relationships of the SPAR-H defined performance shaping factors that affect human error and uses them to assess the overall human error probability of …

Distributed Multi-Label Learning On Apache Spark, Jorge Gonzalez Lopez

Theses and Dissertations

This thesis proposes a series of multi-label learning algorithms for classification and feature selection implemented on the Apache Spark distributed computing model. Five approaches for determining the optimal architecture to speed up multi-label learning methods are presented. These approaches range from local parallelization using threads to distributed computing using independent or shared memory spaces. It is shown that the optimal approach performs hundreds of times faster than the baseline method. Three distributed multi-label k nearest neighbors methods built on top of the Spark architecture are proposed: an exact iterative method that computes pair-wise distances, an approximate tree-based method that indexes …

Effects Of Dynamic Goals On Agent Performance, Nathan R. Ball

Theses and Dissertations

Autonomous systems are increasingly being used for complex tasks in dynamic environments. Robust automation needs to be able to establish its current goal and determine when the goal has changed. In human-machine teams autonomous goal detection is an important component of maintaining shared situational awareness between both parties. This research investigates how different categories of goals affect autonomous change detection in a dynamic environment. In order to accomplish this goal, a set of autonomous agents were developed to perform within an environment with multiple possible goals. The agents perform the environmental task while monitoring for goal changes. The experiment tests …

Efficient Phase Retrieval For Off-Axis Point Spread Functions, Salome Esteban Carrasco

Theses and Dissertations

A novel pairing of phase retrieval tools allows for efficient estimation of pupil phase in optical systems from images of point spread functions (PSFs). The phase retrieval algorithm uses correlation of modeled phase in the focal plane to decouple aberrations that are difficult to identify in complex PSFs. The use of a phase kernel that departs from the Fresnel approximation for off-axis PSFs is a more accurate representation of wavefront phase in finite conjugate imaging. The combination of the approximation and phase correlation algorithm can be more efficient and accurate than generic algorithms.

Methods Of Reverse Engineering A Bitstream For Field Programmable Gate Array Protection, Daniel J. Celebucki

Theses and Dissertations

Field Programmable Gate Arrays (FPGAs) are found in numerous industries including consumer electronics, automotive, military and aerospace, and critical infrastructure. The ability to be reprogrammed as well as large computational power and relatively low price make them a good fit for low-volume applications that cannot justify the Non-Recurring Engineering (NRE) costs associated with producing Application-Specific Integrated Circuits (ASICs). FPGAs however, have seen a variety of security issues stemming from the fact that their configuration files are not inherently protected. This research assesses the feasibility of reverse engineering the bitstream format for a previously unexplored FPGA, as well as the utilization …

Mitigating The Effects Of Boom Occlusion On Automated Aerial Refueling Through Shadow Volumes, Zachary C. Paulson

Theses and Dissertations

In flight refueling of Unmanned Aerial Vehicles (UAVs) is critical to the United States Air Force (USAF). However, the large communication latency between a ground-based operator and his/her remote UAV makes docking with a refueling tanker unsafe. This latency may be mitigated by leveraging a tanker-centric stereo vision system. The vision system observes and computes an approaching receiver's relative position and orientation offering a low-latency, high frequency docking solution. Unfortunately, the boom -- an articulated refueling arm responsible for physically pumping fuel into the receiver -- occludes large portions of the receiver especially as the receiver approaches and docks with …

Stereo Vision: A Comparison Of Synthetic Imagery Vs. Real World Imagery For The Automated Aerial Refueling Problem, Nicholas J. Seydel

Theses and Dissertations

Missions using unmanned aerial vehicles have increased in the past decade. Currently, there is no way to refuel these aircraft. Accomplishing automated aerial refueling can be made possible using the stereo vision system on a tanker. Real world experiments for the automated aerial refueling problem are expensive and time consuming. Currently, simulations performed in a virtual world have shown promising results using computer vision. It is possible to use the virtual world as a substitute environment for the real world. This research compares the performance of stereo vision algorithms on synthetic and real world imagery.

Special Perturbations On The Jetson Tx1 And Tx2 Computers, Tyler M. Moore

Theses and Dissertations

Simplified General Perturbations Number 4 (SGP4) has been the traditional algorithm for performing Orbit Determination (OD) onboard orbiting spacecraft. However, the recent rise of high-performance computers with low Size, Weight, and Power (SWAP) factors has provided the opportunity to use Special Perturbations (SP), a more accurate algorithm to perform onboard OD. This research evaluates the most efficient way to implement SP on NVIDIA’s Jetson TX series of integrated Graphical Processing Units (GPUs). An initial serial version was implemented on the Jetson TX1 and TX2's Central Processing Units (CPUs). The runtimes of the initial version are the benchmark that the runtimes …

Rss-Based Device-Free Passive Detection And Localization Using Home Automation Network Radio Frequencies, Tiffany M. Phan

Theses and Dissertations

This research provided a proof of concept for a device-free passive (DfP) system capable of detecting and localizing a target through exploitation of a home automation network’s radio frequency (RF) signals. The system was developed using Insteon devices with a 915 MHz center frequency. Without developer privileges, limitations of the Insteon technology like no intrinsic received signal strength (RSS) field and silent periods between messages were overcome by using software-defined radios to simulate Insteon devices capable of collecting and reporting RSS, and by creating a message generation script and implementing a calibrated filter threshold to reduce silent periods. Evaluation of …

Integrity Monitoring For Automated Aerial Refueling: A Stereo Vision Approach, Thomas R. Stuart

Theses and Dissertations

Unmanned aerial vehicles (UAVs) increasingly require the capability to y autonomously in close formation including to facilitate automated aerial refueling (AAR). The availability of relative navigation measurements and navigation integrity are essential to autonomous relative navigation. Due to the potential non-availability of the global positioning system (GPS) during military operations, it is highly desirable that relative navigation can be accomplished without the use of GPS. This paper develops two algorithms designed to provide relative navigation measurements solely from a stereo image pair. These algorithms were developed and analyzed in the context of AAR using a stereo camera system modeling that …

Assessment Of Structure From Motion For Reconnaissance Augmentation And Bandwidth Usage Reduction, Jonathan B. Roeber

Theses and Dissertations

Modern militaries rely upon remote image sensors for real-time intelligence. A typical remote system consists of an unmanned aerial vehicle, or UAV, with an attached camera. A video stream is sent from the UAV, through a bandwidth-constrained satellite connection, to an intelligence processing unit. In this research, an upgrade to this method of collection is proposed. A set of synthetic images of a scene captured by a UAV in a virtual environment is sent to a pipeline of computer vision algorithms, collectively known as Structure from Motion. The output of Structure from Motion, a three-dimensional model, is then assessed in …

Target Detection Using Convolutional Neural Networks, Robert P. Loibl

Theses and Dissertations

This research explores the use of Convolutional Neural Networks (CNNs) to classify targets of interest within satellite imagery. Methods were specifically devised for the classification of airports within Landsat-8 scenes. A novel automated dataset generation technique was developed to create labeled datasets from satellite imagery using only coordinate metadata. Using this approach a very large dataset of over 132,000 labeled images was created without human input. This dataset was used to evaluate the effects of color and resolution on airport classification accuracy. Two experiments were run with the first experiment classifying large airports with 96.8% accuracy, and the second classifying …

Smartphone User Privacy Preserving Through Crowdsourcing, Bahman Rashidi

Theses and Dissertations

In current Android architecture, users have to decide whether an app is safe to use or not. Expert users can make savvy decisions to avoid unnecessary private data breach. However, the majority of regular users are not technically capable or do not care to consider privacy implications to make safe decisions. To assist the technically incapable crowd, we propose a permission control framework based on crowdsourcing. At its core, our framework runs new apps under probation mode without granting their permission requests up-front. It provides recommendations on whether to accept or not the permission requests based on decisions from peer …

Estimating The Respiratory Lung Motion Model Using Tensor Decomposition On Displacement Vector Field, Kingston Kang

Theses and Dissertations

Modern big data often emerge as tensors. Standard statistical methods are inadequate to deal with datasets of large volume, high dimensionality, and complex structure. Therefore, it is important to develop algorithms such as low-rank tensor decomposition for data compression, dimensionality reduction, and approximation.

With the advancement in technology, high-dimensional images are becoming ubiquitous in the medical field. In lung radiation therapy, the respiratory motion of the lung introduces variabilities during treatment as the tumor inside the lung is moving, which brings challenges to the precise delivery of radiation to the tumor. Several approaches to quantifying this uncertainty propose using a …

Big Networks: Analysis And Optimal Control, Hung The Nguyen

Theses and Dissertations

The study of networks has seen a tremendous breed of researches due to the explosive spectrum of practical problems that involve networks as the access point. Those problems widely range from detecting functionally correlated proteins in biology to finding people to give discounts and gain maximum popularity of a product in economics. Thus, understanding and further being able to manipulate/control the development and evolution of the networks become critical tasks for network scientists. Despite the vast research effort putting towards these studies, the present state-of-the-arts largely either lack of high quality solutions or require excessive amount of time in real-world …

A Comparative Study Of Cognitive Systems For Learning, Praneetha Mandava

Theses and Dissertations

Learning is the modification of a behavioral tendency by experience. Memory and reasoning are the most important aspects for learning in humans; information is temporarily stored in the short-term memory and processed, compared with existing memories and stored in long-term memory, and can be re-used when needed. One way to describe an organized pattern of thought or behavior and the categories of information along with their relationships is by using schemas. A cognitive script is one form of a schema that evolves over multiple exposures to the same set of stimuli and/or repeated enactment of a particular behavior. This research …

A Trivium-Inspired Pseudorandom Number Generator With A Statistical Comparison To The Randomness Of Securerandom And Trivium, Latoya Niesha Jackson

Theses and Dissertations

A pseudorandom number generator (PRNG) is an algorithm that produces a sequence of numbers which emulates the characteristics of a random sequence. In comparison to its genuine counterpart, PRNGs are considered more suitable for computing devices in that they do not consume a lot of resources (in terms of memory) and their portability; they can also be used on a wide range of devices. Cryptographically Secure PRNGs (CSPRNGs) are the only type of PRNGs suitable for cryptographic applications. They are specially designed to withstand security attacks. In this thesis, we provide descriptions of two CSPRNGs: Trivium, a hardware-based stream cipher …

Network Analytics For The Mirna Regulome And Mirna-Disease Interactions, Joseph Jayakar Nalluri

Theses and Dissertations

miRNAs are non-coding RNAs of approx. 22 nucleotides in length that inhibit gene expression at the post-transcriptional level. By virtue of this gene regulation mechanism, miRNAs play a critical role in several biological processes and patho-physiological conditions, including cancers. miRNA behavior is a result of a multi-level complex interaction network involving miRNA-mRNA, TF-miRNA-gene, and miRNA-chemical interactions; hence the precise patterns through which a miRNA regulates a certain disease(s) are still elusive. Herein, I have developed an integrative genomics methods/pipeline to (i) build a miRNA regulomics and data analytics repository, (ii) create/model these interactions into networks and use optimization techniques, motif …

Application Of Computer Modeling And Simulation Techniques For Optimization Of Factory Floor Operations In Small To Medium- Sized Businesses, Brian P. Romano

Theses and Dissertations

The rationale and motive for this thesis was to prove that no matter the size of a company and its particular value stream, the application of applied computer science principles with a reliance on computer modeling and simulation onto the factory floor process improves efficiencies and throughput through the reduction of downtime and/or process waiting. This thesis research specifically emphasized small businesses of between $2 and $20 million and was purposely limited to factory floor production processes and utilized standardized applied computer science techniques including simulation and modeling, microprocessor based factory floor intelligence devices. The results of this applied technology …

An Exploration Of Rule Clustering In Cellular Automata Rule Spaces, Jordon Huffman

Theses and Dissertations

The study of complex systems examines the global behavior of a system and how the individual parts of the system affect that behavior [1]. The study of complex systems spans across many fields of science like biology, physics, engineering, and computer science. One area of complex systems that has not been fully explored is cellular automata. Since its discovery by John von Neumann, there have been no consistent ways of categorizing similarities between cellular automata rules or collecting similar rules for observation. This thesis introduces an approach to identifying clusters of similar rules and extracting rules from that cluster. Several …

An Arithmetic-Based Deterministic Centroid Initialization Method For The K-Means Clustering Algorithm, Matthew Michael Mayo

Theses and Dissertations

One of the greatest challenges in k-means clustering is positioning the initial cluster centers, or centroids, as close to optimal as possible, and doing so in an amount of time deemed reasonable. Traditional fc-means utilizes a randomization process for initializing these centroids, and poor initialization can lead to increased numbers of required clustering iterations to reach convergence, and a greater overall runtime. This research proposes a simple, arithmetic-based deterministic centroid initialization method which is much faster than randomized initialization. Preliminary experiments suggest that this collection of methods, referred to herein as the sharding centroid initialization algorithm family, often outperforms random …