Browsing by Subject "scaling"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
Item A Delphi Study Assessing Effective Peer Faculty E-Mentoring to Support Scaling Distance Education Programs(2014-04-02) Lewis, Judith HolbrookThis research addressed a gap in the literature regarding the use of e-mentoring as a successful infrastructure mechanism to support educators in delivery of higher education and metrics for its use in scaling online education programs. The methodology applied to this research was a Delphi Study. The Delphi Technique is a qualitative methodology to build a consensus opinion from a panel of experts. This Delphi was based on a series of rounds in which a panel of experts responded to survey questions, each survey item presented as an essentiality statement ranked by a Likert-type scale index from Very Important down to Unimportant. Descriptive statistics were calculated for each survey statement to determine consensus. This study addressed five research questions in the areas of support for distance education faculty: what attributes of an e-mentoring program for faculty engaged in teaching distance education classes lead to perceived effectiveness by coaches and practitioners (terms introduced to describe the mentoring relationship between peers in a community of practice), what formal and informal activities or processes provide for preparation for teaching online, collegiality, and professional development (previously published operationalized factors) (Velez, 2010), and what metrics can be used to determine that e-mentoring has led to increased spread, depth, sustainability, and sense of ownership in distance education, previously published factors for scaling (Coburn, 2003). Based on the Delphi results, the highest consensus means concerned the importance of faculty and administrative support of distance education. For example, the study found high consensus that e-mentoring should be encouraged with release time, coaching should be considered in tenure and promotion decisions, and provision for communication allowances and technical support should be provided for e-mentoring sessions. Training topics of greatest interest included accessibility training, content delivery and teaching modalities, and copyright law and intellectual property expectations. Important metrics included the number of ?formerly coached? practitioners acting as e-mentoring coaches in the future, the number of semester-hours taught, the number of faculty initiating new practices, and faculty acceptance of delivering education online. This study is significant because it researched the use of e-mentoring as a support for faculty in scaling online learning programs in higher education and provided expert evaluation of processes and procedures recommended by faculty to support their effort. It also evaluated metrics to assess the scaling of distance education programs.Item Scaling Up Reinforcement Learning without Sacrificing Optimality by Constraining Exploration(2012-12-05) Mann, Timothy 1984-The purpose of this dissertation is to understand how algorithms can efficiently learn to solve new tasks based on previous experience, instead of being explicitly programmed with a solution for each task that we want it to solve. Here a task is a series of decisions, such as a robot vacuum deciding which room to clean next or an intelligent car deciding to stop at a traffic light. In such a case, state-of-the-art learning algorithms are difficult to employ in practice because they often make thou- sands of mistakes before reliably solving a task. However, humans learn solutions to novel tasks, often making fewer mistakes, which suggests that efficient learning algorithms may exist. One advantage that humans have over state- of-the-art learning algorithms is that, while learning a new task, humans can apply knowledge gained from previously solved tasks. The central hypothesis investigated by this dissertation is that learning algorithms can solve new tasks more efficiently when they take into consideration knowledge learned from solving previous tasks. Al- though this hypothesis may appear to be obviously true, what knowledge to use and how to apply that knowledge to new tasks is a challenging, open research problem. I investigate this hypothesis in three ways. First, I developed a new learning algorithm that is able to use prior knowledge to constrain the exploration space. Second, I extended a powerful theoretical framework in machine learning, called Probably Approximately Correct, so that I can formally compare the efficiency of algorithms that solve only a single task to algorithms that consider knowledge from previously solved tasks. With this framework, I found sufficient conditions for using knowledge from previous tasks to improve efficiency of learning to solve new tasks and also identified conditions where transferring knowledge may impede learning. I present situations where transfer learning can be used to intelligently constrain the exploration space so that optimality loss can be minimized. Finally, I tested the efficiency of my algorithms in various experimental domains. These theoretical and empirical results provide support for my central hypothesis. The theory and experiments of this dissertation provide a deeper understanding of what makes a learning algorithm efficient so that it can be widely used in practice. Finally, these results also contribute the general goal of creating autonomous machines that can be reliably employed to solve complex tasks.Item Session 2E | Improving Image Processing Through Iteration and Automation(Texas Digital Library, 2021-05-25) Jones, Jerrell; Watkins, SeanUH Libraries has been building a robust and flexible digital collections ecosystem since 2016. In 2020, UH Libraries launched its digital collections ecosystem that supports efficient digital collections management, effective digital preservation, and integration with library systems. These goals encompassed new tooling in the image processing workflow to help manage the demand at scale and facilitate the production of high-quality output. UH Libraries has implemented command-line based scripts to address inefficient legacy workflows. We will give examples of what workflows were created, tools developed and utilized for key processes, the progression of these tools through the digital projects agile team, and automation developed into UH Libraries digital projects management application, Mason. We will also examine some of the transformative knowledge gained during these iterative processes that contribute to a more efficient production environment. As UH Libraries continues its migration of collections into a new repository, these tools continue to be applied and improvements are being made to the image processing workflow and its tools.Item Zirconium-doped tantalum oxide high-k gate dielectric films(Texas A&M University, 2005-02-17) Tewg, Jun-YenA new high-k dielectric material, i.e., zirconium-doped tantalum oxide (Zr-doped TaOx), in the form of a sputter-deposited thin film with a thickness range of 5-100 nm, has been studied. Important applications of this new dielectric material include the gate dielectric layer for the next generation metal-oxide-semiconductor field effect transistor (MOSFET). Due to the aggressive device scaling in ultra-large-scale integrated circuitry (ULSI), the ultra-thin conventional gate oxide (SiO2) is unacceptable for many practical reasons. By replacing the SiO2 layer with a high dielectric constant material (high-k), many of the problems can be solved. In this study, a novel high-k dielectric thin film, i.e., TaOx doped with Zr, was deposited and studied. The film?s electrical, chemical, and structural properties were investigated experimentally. The Zr dopant concentration and the thermal treatment condition were studied with respect to gas composition, pressure, temperature, and annealing time. Interface layer formation and properties were studied with or without an inserted thin tantalum nitride (TaNx) layer. The gate electrode material influence on the dielectric properties was also investigated. Four types of gate materials, i.e., aluminum (Al), molybdenum (Mo), molybdenum nitride (MoN), and tungsten nitride (WN), were used in this study. The films were analyzed with ESCA, XRD, SIMS, and TEM. Films were made into MOS capacitors and characterized using I-V and C-V curves. Many promising results were obtained using this kind of high-k film. It is potentially applicable to future MOS devices.