Digital Commons Network^™

Deep Representation Learning For Clustering And Domain Adaptation, Mohsen Kheirandishfard

Computer Science and Engineering Dissertations - Archive

Representation learning is a fundamental task in the area of machine learning which can significantly influence the performance of the algorithms used in various applications. The main goal of this task is to capture the relationships between the input data and learn feature representations that contain the most useful information of the original data. Such representations can be further leveraged in many machine learning applications such as clustering, natural language analysis, recommender systems, etc. In this dissertation, we first present a theoretical framework for solving a broad class of non-convex optimization problems. The proposed method is applicable to various tasks …

Multivariate Time Series Pattern Recognition Using Machine Learning And Deep Learning Methods, Sai Abhishek Devar

Industrial, Manufacturing, and Systems Theses - Archive

In this research work, we have implemented machine learning & deep-learning algorithms on real-time multivariate time series datasets in the manufacturing & health care fields. The research work is organized in two case-studies. The case study-1 is about rare event classification in multivariate time series in a pulp and paper manufacturing industry, data was collected of multiple sensors at each stage of production line, the data contains a rare event of paper break that commonly occurs in the industry. For preprocessing we have implemented sliding window approach for calculating first order difference method to capture the variation in the data …

Boosting: Why You Can Use The Hp Filter, Peter C.B. Phillips, Zhentao Shi

Cowles Foundation Discussion Papers

The Hodrick-Prescott (HP) filter is one of the most widely used econometric methods in applied macroeconomic research. The technique is nonparametric and seeks to decompose a time series into a trend and a cyclical component unaided by economic theory or prior trend specification. Like all nonparametric methods, the HP filter depends critically on a tuning parameter that controls the degree of smoothing. Yet in contrast to modern nonparametric methods and applied work with these procedures, empirical practice with the HP filter almost universally relies on standard settings for the tuning parameter that have been suggested largely by experimentation with macroeconomic …

Domain Adaptation In Unmanned Aerial Vehicles Landing Using Reinforcement Learning, Pedro Lucas Franca Albuquerque

School of Computing: Dissertations, Theses, and Student Research

Landing an unmanned aerial vehicle (UAV) on a moving platform is a challenging task that often requires exact models of the UAV dynamics, platform characteristics, and environmental conditions. In this thesis, we present and investigate three different machine learning approaches with varying levels of domain knowledge: dynamics randomization, universal policy with system identification, and reinforcement learning with no parameter variation. We first train the policies in simulation, then perform experiments both in simulation, making variations of the system dynamics with wind and friction coefficient, then perform experiments in a real robot system with wind variation. We initially expected that providing …

Toward Self-Reconfigurable Parametric Systems: Reinforcement Learning Approach, Ting-Yu Mu

Dissertations

For the ongoing advancement of the fields of Information Technology (IT) and Computer Science, machine learning-based approaches are utilized in different ways in order to solve the problems that belong to the Nondeterministic Polynomial time (NP)-hard complexity class or to approximate the problems if there is no known efficient way to find a solution. Problems that determine the proper set of reconfigurable parameters of parametric systems to obtain the near optimal performance are typically classified as NP-hard problems with no efficient mathematical models to obtain the best solutions. This body of work aims to advance the knowledge of machine learning …

Characterizing Dryland Ecosystems Using Remote Sensing And Dynamic Global Vegetation Modeling, Abdolhamid Dashtiahangar

Boise State University Theses and Dissertations

Drylands include all terrestrial regions where the production of crops, forage, wood and other ecosystem services are limited by water. These ecosystems cover approximately 40% of the earth terrestrial surface and accommodate more than 2 billion people (Millennium Ecosystem Assessment, 2005). Moreover, the interannual variability of the global carbon budget is strongly regulated by vegetation dynamics in drylands. Understanding the dynamics of such ecosystems is significant for assessing the potential for and impacts of natural or anthropogenic disturbances and mitigation planning, and a necessary step toward enhancing the economic and social well-being of dryland communities in a sustainable manner (Global …

Comparison Of Rl Algorithms For Learning To Learn Problems, Adolfo Gonzalez Iii

Theses and Dissertations

Machine learning has been applied to many different problems successfully due to the expressiveness of neural networks and simplicity of first order optimization algorithms. The latter being a vital piece needed for training large neural networks efficiently. Many of these algorithms were produced with behavior produced by experiments and intuition. An interesting question that comes to mind is that rather than observing and then designing algorithms with beneficial behaviors, can these algorithms be learned through a reinforcement learning by modeling optimization as a game. This paper explores several reinforcement learning algorithms which are applied to learn policies suited for optimization.

Materials Prediction Using High-Throughput And Machine Learning Techniques, Chandramouli Nyshadham

Theses and Dissertations

Predicting new materials through virtually screening a large number of hypothetical materials using supercomputers has enabled materials discovery at an accelerated pace. However, the innumerable number of possible hypothetical materials necessitates the development of faster computational methods for speedier screening of materials reducing the time of discovery. In this thesis, I aim to understand and apply two computational methods for materials prediction. The first method deals with a computational high-throughput study of superalloys. Superalloys are materials which exhibit high-temperature strength. A combinatorial high-throughput search across 2224 ternary alloy systems revealed 102 potential superalloys of which 37 are brand new, all …

Machine Learning Predictions Electronic Couplings For Charge Transport Calculations Of P3ht, Evan D. Miller, Matthew L. Jones, Mike M. Henry, Bryan Stanfill, Eric Jankowski

Materials Science and Engineering Faculty Publications and Presentations

The purpose of this work is to lower the computational cost of predicting charge mobilities in organic semiconductors, which will benefit the screening of candidates for inexpensive solar power generation. We characterize efforts to minimize the number of expensive quantum chemical calculations we perform by training machines to predict electronic couplings between monomers of poly-(3-hexylthiophene). We test five machine learning techniques and identify random forests as the most accurate, information-dense, and easy-to-implement approach for this problem, achieving mean-absolute-error of 0.02 [× 1.6 × 10⁻¹⁹ J], R² = 0.986, predicting electronic couplings 390 times faster than quantum chemical calculations, …

Sleep Stage Prediction With Raw Acceleration And Photoplethysmography Heart Rate Data Derived From A Consumer Wearable Device, Olivia Walch, Yitong Huang, Daniel Forger, Cathy Goldstein

Mathematics Sciences: Faculty Publications

Wearable, multisensor, consumer devices that estimate sleep are now commonplace, but the algorithms used by these devices to score sleep are not open source, and the raw sensor data is rarely accessible for external use. As a result, these devices are limited in their usefulness for clinical and research applications, despite holding much promise. We used a mobile application of our own creation to collect raw acceleration data and heart rate from the Apple Watch worn by participants undergoing polysomnography, as well as during the ambulatory period preceding in lab testing. Using this data, we compared the contributions of multiple …

Mdt Geolocation Through Machine Learning: Evaluation Of Supervised Regression Ml Algorithms, Aria Canadell Solana

Theses and Dissertations

Minimizing Drive Test is a statistical protocol used to evaluate the network performance. It provides several benefits with respect to traditional drive test analysis; however, multiple inconveniences exist that prevent cell companies from precisely retrieving most of the locations of these reports. . MATLAB and Jupyter Notebook were used to prepare the data and create the models. Multiple supervised regression algorithms were tested and evaluated. The best predictions were obtained from the K-Nearest Neighbor algorithm with one ‘k’ and distance-weighted predictions. The UE geolocation was predicted with a median accuracy of 5.42 meters, a mean error of 61.62 meters, and …

Poly-Gan: A Multi-Conditioned Gan For Multiple Tasks, Nilesh Pandey

Theses

We present Poly-GAN, a novel conditional GAN architecture that is motivated by different Image generation and manipulation applications like Fashion Synthesis, an application where garments are automatically placed on images of human models at an arbitrary pose, image inpainting, an application where we try to recover a damaged image using the edges or a rough sketch of the image. While different applications use different GAN setup for image generation, we propose only one architecture for multiple applications with little to no change in the pipeline. Poly-GAN allows conditioning on multiple inputs and is suitable for many different tasks. Our novel …

Countering Cybersecurity Vulnerabilities In The Power System, Fengli Zhang

Graduate Theses and Dissertations

Security vulnerabilities in software pose an important threat to power grid security, which can be exploited by attackers if not properly addressed. Every month, many vulnerabilities are discovered and all the vulnerabilities must be remediated in a timely manner to reduce the chance of being exploited by attackers. In current practice, security operators have to manually analyze each vulnerability present in their assets and determine the remediation actions in a short time period, which involves a tremendous amount of human resources for electric utilities. To solve this problem, we propose a machine learning-based automation framework to automate vulnerability analysis and …

Pixel-Level Deep Multi-Dimensional Embeddings For Homogeneous Multiple Object Tracking, Mateusz Mittek

Department of Electrical and Computer Engineering: Dissertations, Theses, and Student Research

The goal of Multiple Object Tracking (MOT) is to locate multiple objects and keep track of their individual identities and trajectories given a sequence of (video) frames. A popular approach to MOT is tracking by detection consisting of two processing components: detection (identification of objects of interest in individual frames) and data association (connecting data from multiple frames). This work addresses the detection component by introducing a method based on semantic instance segmentation, i.e., assigning labels to all visible pixels such that they are unique among different instances. Modern tracking methods often built around Convolutional Neural Networks (CNNs) and additional, …

Augmenting Education: Ethical Considerations For Incorporating Artificial Intelligence In Education, Dana Remian

Instructional Design Capstones Collection

Artificial intelligence (AI) has existed in theory and practice for decades, but applications have been relatively limited in most domains. Recent developments in AI and computing have placed AI-enhanced applications in various industries and a growing number of consumer products. AI platforms and services aimed at enhancing educational outcomes and taking over administrative tasks are becoming more prevalent and appearing in more and more classrooms and offices. Conversations about the disruption and ethical concerns created by AI are occurring in many fields. The development of the technology threatens to outpace academic discussion of its utility and pitfalls in education, however. …

“Where’S The I-O?” Artificial Intelligence And Machine Learning In Talent Management Systems, Manuel F. Gonzalez, John F. Capman, Frederick L. Oswald, Evan R. Theys, David L. Tomczak

Personnel Assessment and Decisions

Artificial intelligence (AI) and machine learning (ML) have seen widespread adoption by organizations seeking to identify and hire high-quality job applicants. Yet the volume, variety, and velocity of professional involvement among I-O psychologists remains relatively limited when it comes to developing and evaluating AI/ML applications for talent assessment and selection. Furthermore, there is a paucity of empirical research that investigates the reliability, validity, and fairness of AI/ML tools in organizational contexts. To stimulate future involvement and research, we share our review and perspective on the current state of AI/ML in talent assessment as well as its benefits and potential pitfalls; …

Virtual Wrap-Up Presentation: Digital Libraries, Intelligent Data Analytics, And Augmented Description, Elizabeth Lorang, Leen-Kiat Soh, Yi Liu, Chulwoo Pack

School of Computing: Conference and Workshop Papers

Includes framing, overview, and discussion of the explorations pursued as part of the Digital Libraries, Intelligent Data Analytics, and Augmented Description demonstration project, pursued by members of the Aida digital libraries research team at the University of Nebraska-Lincoln through a research services contract with the Library of Congress. This presentation covered: Aida research team and background for the demonstration project; broad outlines of “Digital Libraries, Intelligent Data Analytics, and Augmented Description”; what changed for us as a research team over the collaboration and why; deliverables of our work; thoughts toward “What next”; and deep-dives into the explorations. The machine learning …

Word Importance Modeling To Enhance Captions Generated By Automatic Speech Recognition For Deaf And Hard Of Hearing Users, Sushant Kafle

Theses

People who are deaf or hard-of-hearing (DHH) benefit from sign-language interpreting or live-captioning (with a human transcriptionist), to access spoken information. However, such services are not legally required, affordable, nor available in many settings, e.g., impromptu small-group meetings in the workplace or online video content that has not been professionally captioned. As Automatic Speech Recognition (ASR) systems improve in accuracy and speed, it is natural to investigate the use of these systems to assist DHH users in a variety of tasks. But, ASR systems are still not perfect, especially in realistic conversational settings, leading to the issue of trust and …

Using Uncertainty To Interpret Supervised Machine Learning Predictions, Michael C. Darling

Electrical and Computer Engineering ETDs

Traditionally, machine learning models are assessed using methods that estimate an average performance against samples drawn from a particular distribution. Examples include the use of cross-validation or hold0out to estimate classification error, F-score, precision, and recall.

While these measures provide valuable information, they do not tell us a model's certainty relative to particular regions of the input space. Typically there are regions where the model can differentiate the classes with certainty, and regions where the model is much less certain about its predictions.

In this dissertation we explore numerous approaches for quantifying uncertainty in the individual predictions made by supervised …

Detecting Digitally Forged Faces In Online Videos, Neilesh Sambhu

USF Tampa Graduate Theses and Dissertations

We use Rossler’s FaceForensics dataset of 1004 online videos and their corresponding forged counterparts [1] to investigate the ability to distinguish digitally forged facial images from original images automatically with deep learning. The proposed convolutional neural network is much smaller than the current state-of-the-art solutions. Nevertheless, the network maintains a high level of accuracy (99.6%), all while using the entire FaceForensics dataset and not including any temporal information. We implement majority voting and show the impact on accuracy (99.67%), where only 1 video of 300 is misclassified. We examine why the model misclassified this one video. In terms of tuning …

Deep Neural Network Based Iterative Self-Taught Learning On Text Mining, Xiangwen Liu

Theses and Dissertations

[Motivation] Machine learning and deep learning models is the main methodology for discovering hidden knowledge of data. Data plays a significant role in the training process. However, lack of data especially labeled data is precious due to accessibility and high cost. How to use the abundant unlabeled data to enhance the downstream learning process on labeled data is an urgent task in machine learning and pattern recognition. [Methods] An STL (Self-taught learning) model is proposed in this dissertation, which can successfully learn features from downloaded unlabeled text and enhance the downstream tasks on labeled text. In addition, DLI-IT (Drug labeling …

Study Of Human Hand-Eye Coordination Using Machine Learning Techniques In A Virtual Reality Setup, Kamran Binaee

Theses

Theories of visually guided action are characterized as closed-loop control in the presence of reliable sources of visual information, and predictive control to compensate for visuomotor delay and temporary occlusion. However, prediction is not well understood. To investigate, a series of studies was designed to characterize the role of predictive strategies in humans as they perform visually guided actions, and to guide the development of computational models that capture these strategies. During data collection, subjects were immersed in a virtual reality (VR) system and were tasked with using a paddle to intercept a virtual ball. To force subjects into a …

Collaborating On Machine Reading: Training Algorithms To Read Complex Collections, Carrie M. Pirmann, Brian R. King, Bhagawat Acharya, Katherine M. Faull

Bucknell University Digital Scholarship Conference

Interdisciplinary collaboration between two faculty members in the humanities and computer science, a research librarian, and an undergraduate student has led to remarkable results in an ongoing international DH research project that has at its core 18th century manuscripts. The corpus stems from a vast collection of archival materials held by the Moravian Church in the UK, Germany, and the US. The number of pages to be transcribed, differences in handwriting styles, paper quality, and original language pose enormous problems for the feasibility of human transcription. This presentation will review the hypothesis, process, and findings of a summer research project …

Machine Learning To Support An Interactive Theorem Prover, Salman Haider, Andy Le, Echo Wu, Brian T. Howard

Annual Student Research Poster Session

An Interactive Theorem Prover (ITP) is a computer program that can assist a human in creating a proof of a mathematical theorem or the correctness of a piece of software. At each step in the proof, the human has the computer apply a chosen "tactic" to attempt to make progress toward the goal. We are exploring the possibility of using Machine Learning (ML) to assist in this tactic selection. Building on recent work at UC Berkeley and Princeton, where they adapted the Coq ITP so that it could interface with ML libraries via the Python scripting language, we have been …

Adaptive Feature Engineering Modeling For Ultrasound Image Classification For Decision Support, Hatwib Mugasa

Doctoral Dissertations

Ultrasonography is considered a relatively safe option for the diagnosis of benign and malignant cancer lesions due to the low-energy sound waves used. However, the visual interpretation of the ultrasound images is time-consuming and usually has high false alerts due to speckle noise. Improved methods of collection image-based data have been proposed to reduce noise in the images; however, this has proved not to solve the problem due to the complex nature of images and the exponential growth of biomedical datasets. Secondly, the target class in real-world biomedical datasets, that is the focus of interest of a biopsy, is usually …

Demonstration Of Visible And Near Infrared Raman Spectrometers And Improved Matched Filter Model For Analysis Of Combined Raman Signals, Alexander Matthew Atkinson

Electrical & Computer Engineering Theses & Dissertations

Raman spectroscopy is a powerful analysis technique that has found applications in fields such as analytical chemistry, planetary sciences, and medical diagnostics. Recent studies have shown that analysis of Raman spectral profiles can be greatly assisted by use of computational models with achievements including high accuracy pure sample classification with imbalanced data sets and detection of ideal sample deviations for pharmaceutical quality control. The adoption of automated methods is a necessary step in streamlining the analysis process as Raman hardware becomes more advanced. Due to limits in the architectures of current machine learning based Raman classification models, transfer from pure …

Development Of A National-Scale Big Data Analytics Pipeline To Study The Potential Impacts Of Flooding On Critical Infrastructures And Communities, Nattapon Donratanapat

Theses and Dissertations

With the rapid development of the Internet of Things (IoT) and Big data infrastructure, crowdsourcing techniques have emerged to facilitate data processing and problem solving particularly for flood emergences purposes. A Flood Analytics Information System (FAIS) has been developed as a Python Web application to gather Big data from multiple servers and analyze flooding impacts during historical and real-time events. The application is smartly designed to integrate crowd intelligence, machine learning (ML), and natural language processing of tweets to provide flood warning with the aim to improve situational awareness for flood risk management and decision making. FAIS allows the user …

Machine Learning Based Ultra High Carbon Steel Image Segmentation, Sumith Kuttiyil Suresh

Theses and Dissertations

Mechanical and structural properties of ultra-high carbon steel are determined by their microstructures composed of constituents such as pearlite and spheroidites. Locating micro constituents and quantitatively measuring its presence is key for material researchers to study the physical properties of the carbon steel materials. This micrograph analysis is currently done manually and subjectively by material scientists, which is tedious and time-consuming. Here we propose to apply the image segmentation algorithm called U-Net to achieve automated labeling of steel microstructures on a subset of ultra- high carbon steel image dataset containing pearlite and spheroidite as the primary micro constituents. Our work …

Discovery Of Materials Through Applied Machine Learning, Travis Williams

Theses and Dissertations

Advances in artificial intelligence technology, specifically machine learning, have cre- ated opportunities in the material sciences to accelerate material discovery and gain fundamental understanding of the interaction between certain the constituent ele- ments of a material and the properties expressed by that material. Application of machine learning to experimental materials discovery is slow due to the monetary and temporal cost of experimental data, but parallel techniques such as continuous com- positional gradients or high-throughput characterization setups are capable of gener- ating larger amounts of data than the typical experimental process, and therefore are suitable for combination with machine learning. A …

Algorithmic Advertising Discrimination, Joseph Blass

Northwestern University Law Review

The ability of social media companies to precisely target advertisements to individual users based on those users’ characteristics is changing how job opportunities are advertised. Companies like Facebook use machine learning to place their ads, and machine learning systems present risks of discrimination, which current legal doctrines are not designed to deal with. This Note will explain why it is difficult to ensure such systems do not learn discriminatory functions and why it is hard to discern what they have learned as long as they appear to be performing well on their assigned task. This Note then shows how litigation …