Indiana University-Purdue University Indianapolis (IUPUI) / Current research in deep learning is primarily focused on using Python as a support language. Go, an emerging language, that has many benefits including native support for concurrency has seen a rise in adoption over the past few years. However, this language is not widely used to develop learning models due to the lack of supporting libraries and frameworks for model development. In this thesis, the use of Go for the development of neural network models in general and convolution neural networks is explored. The proposed study is based on a Go-CUDA implementation of neural network models called GoCuNets. This implementation is then compared to a Go-CPU deep learning implementation that takes advantage of Go's built in concurrency called ConvNetGo. A comparison of these two implementations shows a significant performance gain when using GoCuNets compared to ConvNetGo.

<p>  </p> <p>The abundance of heterogeneous data produced and collected each day via multimodal sources may contain hidden events of interest, but in order to extract them the streams of data need to be analyzed with appropriate algorithms, so these events are presented to the end user at the right moment and at the right time. This dissertation proposes a series of algorithms that shape a comprehensive framework for situational knowledge on demand to address this problem. The framework consists of several modules and approaches, each of them is presented in a separate chapter: I begin with video data analysis in streaming video and video at rest for enhanced object detection of real-life surveillance video. For detecting the rare events of interest, I develop a semantic video analysis algorithm which uses an overlay knowledge graph and a semantical network. I show that the usage of the external knowledge for understanding the semantic analysis outperforms other techniques such as transfer learning. </p> <p>The semantical outliers can be used further for improving the algorithm of detecting new objects in the stream of different modalities. I extend the framework with additional modules for natural language data and apply the extended version of the semantic analysis algorithm to define the events of interest from multimodal streaming data. I present a way of combining several feature extractors which can be extended to multiple heterogeneous streams of data in order to efficiently fuse the data based on its semantical similarity, and then show how the serverless architecture of the framework outperforms conventional cloud software architecture. </p> <p>Besides detecting the rare and semantically incompatible events, the semantic analysis can be used for improving the neural networks performance with the data augmentation. The algorithm presented for augmenting the data with the potentially novel objects to circumvent the data drift problem uses the knowledge graph and generative adversarial networks to present the objects to augment the training datasets for supervised learning. I extend the presented framework with a pipeline for generating synthetic novelties to improve the performance of feature extractors and provide the empirical evaluation of the developed method.</p>

The perceptions of Southern and East African learners of their experience in a continuing education were examined using select dimensions of Khan’s e-learning evaluation framework as a theoretical one. This framework provided an evaluative frame in which to examine attributes of e-learning in the context of globalization, internationalization, and massification of education. This research provided a baseline of student perception about the utility and suitability of online education to meet the growing international demands for education. Technology has played a role in the delivery of education, such as that of recent emerging technologies, notably social media. Data included learner reflections and survey responses. A qualitative analysis methodology was used to analyze responses to the research questions. The findings from this research provided insight about student perceptions, engagement, and acceptance of social media as complementary tools in the delivery of educational programming for Southern and East African learners taking blended learning courses from Canadian universities with a proficient African facilitator. These findings permitted some understanding of the technological, social, cultural, and motivational challenges (Dimensions of Kahn’s framework) associated with distance education for African adult learners to permit more nuanced course design and delivery.

Zugl.: Paderborn, Univ., Diss., 2008

D.Ed. / The research study focused on how teachers respond to policy demands given the challenges of rural education and dominating policy discourses and how their professional learning is shaped by these conditions and limitations. In particular I wanted to learn and document what and how teachers learn as they engage in Integrated Quality Management Systems. This study was conducted because there is an outcry of poor examinations results in some rural schools, which could be linked to poor implementation of policies. To respond to this outcry the Teacher Learning Framework which could be used by teachers was developed. The investigation into teacher learning processes when they engage in policy implementation was designed as an interpretive ethnographic study and was conducted in a disadvantaged rural primary school in Ladysmith, in KwaZulu Natal, in South Africa. The study involved five teachers and it was undertaken as an action research study in two cycles. The purpose of cycle one was to verify how teachers presented lessons in terms of policy requirements and guidelines, and authenticate how they learn in the process about their teaching, learners, themselves, and policy requirements. Lessons conducted by teachers were observed and the reflections were also made by two teachers. This was done also as a strategy to gather data. In cycle one, data revealed that teachers were not in a position to implement what they planned to implement. Teachers gave themselves time to plan for cycle two in order to improve on the shortcomings in cycle one. Teachers appeared to have prepared their lessons well, involving learners. Some teachers indicated clear lessons steps which were followed in the classroom. Even before they went to teach in class, they discussed their lessons plans which contributed to the improvement in cycle two. As a process of data gathering, five teachers were interviewed. Interviews were transcribed and analyzed. As a second level of data analysis which was towards developing the Teacher Professional Learning Framework, all the data types were visited with the aim of extracting and discovering codes and categories across data types. These codes and categories were used in a more thorough grounded analysis to formulate a Teacher Professional Learning Framework.

The theoretical aspects of Architectural Management (AM) have been widely researched from a variety of international perspectives through the published research work of the CIB W096 Working Group. There is much less research, however, covering the transfer of these theoretical aspects into professional practice. There is a lack of a holistic approach towards defining AM, explaining what it constitutes, its deliverables to its users, whether there is a need for architectural managers, and if so, what are their qualifications. Similarly, there is a lack of research concerning the opinions of those outside the CIB W096 community with regard to Architectural Management. This research has contributed to the theory and practice of AM by conducting a combination of inductive/deductive, exploratory/explanatory, and qualitative/quantitative approaches in order to understand the exact meaning of AM; capture and detail its components; and to validate all these issues by examining the professional opinions of two groups: researchers and practitioners. The primary focus of this research was answering the question as to how AM can be transferred successfully from theory into practice. In order to answer this question, the research was divided into five consecutive phases. Firstly, reviewing the literature helped establish a solid theoretical background for the research, and it helped in highlighting the major gaps in knowledge associated with AM. Subsequently, and in response to the shortage of information within the AM literature, a preliminary study was found to be a useful source for gathering information about the meaning, components, benefits, strategies and requirements of AM, and the need for architectural managers. Then, both the data extracted from the literature and the data obtained through the preliminary study were analysed and combined, generating a framework for transferring AM from theory to practice using the grounded theory methodology. Then the AM framework testing process was conducted in three stages through a workshop, interviews, and questionnaire survey examining the professional perspectives of architectural researchers (within and outside the CIB W096) and the leaders of UK architectural firms. The final stage involved discussing and synthesising the data obtained through the entire course of this research and generating conclusions. During its testing and after refinement, the newly proposed AM framework proved its practicality and usefulness for transferring Architectural Management from theory to practice. The findings indicated that the decision to adopt AM should be taken at the firm's strategic management level; and there is a need for a facilitator expert in design, management and construction to lead the successful adoption and application of AM. Similarly, the findings revealed that adopting AM in practice has the capability to increase a firm's competitiveness. However, this requires effective communication, collaboration, and knowledge sharing within the firm's internal and external environments. Similarly, the findings indicated the crucial role of basic and vocational architecture education in spreading the concept of AM and assuring its successful application.

Current research in deep learning is primarily focused on using Python as a support language. Go, an emerging language, that has many benefits including native support for concurrency has seen a rise in adoption over the past few years. However, this language is not widely used to develop learning models due to the lack of supporting libraries and frameworks for model development. In this thesis, the use of Go for the development of neural network models in general and convolution neural networks is explored. The proposed study is based on a Go-CUDA implementation of neural network models called GoCuNets. This implementation is then compared to a Go-CPU deep learning implementation that takes advantage of Go's built in concurrency called ConvNetGo. A comparison of these two implementations shows a significant performance gain when using GoCuNets compared to ConvNetGo.<br>

Through the years, educational theorists and researchers have been interested in a possible relationship between teachers' effectiveness and their beliefs and expectations. Three concepts underpinned this work: teacher effectiveness, cultural diversity, and teachers' beliefs and expectations. The premise of the study was that the beliefs and expectations of effective secondary choral teachers are related to the social-cultural contexts in which they teach. The study implemented critical discourse analysis as the theoretical framework and the in-depth phenomenological long interview for data collection. Three secondary choral teachers were selected to participate in the study based on the researcher's criteria. The study revealed how each teacher conceptualized student cultural diversity during the teaching experience. Teacher beliefs about effective teaching in culturally diverse settings were described as developing over time in phases along a continuum. The study also confirmed that teachers' beliefs about students can be changed through experiences and reflection. The study revealed effective teachers focused on three different types of expectations in the teaching and learning context and affirmed diverse cultural identities and backgrounds. Recommendations included the development of stronger mentorship programs to increase effective teaching strategies for the secondary choral classroom. The findings of this study support my previous work, which introduces a sequential learning framework for teaching music in culturally diverse schools.

No description available.

This research explored the factors that influence whether Learning Support Assistants (LSAs), who support the learning of pupils with very complex learning needs, are equipped with the necessary skills to help these students learn. A significant feature in the delivery of educational support to pupils with low incidence needs is that it is LSAs who provide the majority of educational instruction and, furthermore, these paraeducators work with only limited supervision from class teachers. It therefore follows that if high quality educational assistance is to be available to pupils, then those undertaking the majority of this support need to have appropriate pedagogical knowledge. I explored this complex subject by undertaking a workplace ethnography at a single special school over the course of an academic year. In my ethnography, I used a number of data collection methods, including semi-structured interviews with teachers, LSAs and therapists, as well as participant and non-participant observations. The limited literature on the topic of LSA learning concludes that they are not being provided with the learning opportunities they need for the demanding work they do. However, because of the scarcity of information in the literature pertaining to the development of this important group of the educational workforce, I needed to look to the literature on organisational learning more broadly to inform my work and relate it to my own area of interest. The lack of priority given to non-formal learning was a key theme across these texts, with opportunities for participative learning emerging as an important feature. However, this aspect alone could not explain the reasons why these paraeducators did not have access to learning provision appropriate to their needs and so I looked to other theories of workplace learning. The conceptual and analytical approach of the Working as Learning Framework (WALF) (Felstead, et al., 2009), with its incorporation of the concepts of systems of production, discretion, and learning environments and territories, offered me the opportunity to scrutinise the situation regarding the learning and development of LSAs from a wide perspective. Through my adoption of the WALF I have identified the influences that shaped the workplace learning environment of the LSAs at the special school. In doing so, I have added to the limited research on this important and yet inadequately understood group. Although my study focused on one special school this research can inform how the abilities of paraeducators in different educational settings can be developed, because of my adoption of the theoretical standpoint of the WALF. Furthermore, by applying the WALF, my thesis has utilised workplace learning theory to make an important intellectual contribution to the discussion about how high quality educational provision can be delivered. My thesis is also apposite, because the number of children with more complex learning difficulties is increasing and these individuals require specialist paraeducators to be appropriately equipped to meet their needs, whether they are being educated in special or mainstream schools.