Proposing the deep dynamic Bayesian network as a future computer based medical system

The development of new learning models has been of great importance throughout recent years, with a focus on creating advances in the area of deep learning. Deep learning was first noted in 2006, and has since become a major area of research in a number of disciplines. This paper will delve into the area of deep learning to present its current limitations and provide a new idea for a fully integrated deep and dynamic probabilistic system. The new model will be applicable to a vast number of areas initially focusing on applications into medical image analysis with an overall goal of utilising this approach for prediction purposes in computer based medical systems.

Machine learning and data decompositions for complex networked dynamical systems

Thesis (Ph.D.)--University of Washington, 2016-08

Deep convolutional neural networks for semantic video object segmentation

In this thesis, we propose to infer pixel-level labelling in video by utilising only object category information, exploiting the intrinsic structure of video data. Our motivation is the observation that image-level labels are much more easily to be acquired than pixel-level labels, and it is natural to find a link between the image level recognition and pixel level classification in video data, which would transfer learned recognition models from one domain to the other one. To this end, this thesis proposes two domain adaptation approaches to adapt the deep convolutional neural network (CNN) image recognition model trained from labelled image data to the target domain exploiting both semantic evidence learned from CNN, and the intrinsic structures of unlabelled video data. Our proposed approaches explicitly model and compensate for the domain adaptation from the source domain to the target domain which in turn underpins a robust semantic object segmentation method for natural videos. We demonstrate the superior performance of our methods by presenting extensive evaluations on challenging datasets comparing with the state-of-the-art methods.

Learning to construct our identities over the life course: a study with lesbian, gay, bisexual and transgender adults in Scotland

To date, adult educational research has had a limited focus on lesbian, gay, bisexual and transgendered (LGBT) adults and the learning processes in which they engage across the life course. Adopting a biographical and life history methodology, this study aimed to critically explore the potentially distinctive nature and impact of how, when and where LGBT adults learn to construct their identities over their lives. In-depth, semi-structured interviews, dialogue and discussion with LGBT individuals and groups provided rich narratives that reflect shifting, diverse and multiple ways of identifying and living as LGBT. Participants engage in learning in unique ways that play a significant role in the construction and expression of such identities, that in turn influence how, when and where learning happens. Framed largely by complex heteronormative forces, learning can have a negative, distortive impact that deeply troubles any balanced, positive sense of being LGBT, leading to self- censoring, alienation and in some cases, hopelessness. However, learning is also more positively experiential, critically reflective, inventive and queer in nature. This can transform how participants understand their sexual identities and the lifewide spaces in which they learn, engendering agency and resilience. Intersectional perspectives reveal learning that participants struggle with, but can reconcile the disjuncture between evolving LGBT and other myriad identities as parents, Christians, teachers, nurses, academics, activists and retirees. The study’s main contributions lie in three areas. A focus on LGBT experience can contribute to the creation of new opportunities to develop intergenerational learning processes. The study also extends the possibilities for greater criticality in older adult education theory, research and practice, based on the continued, rich learning in which participants engage post-work and in later life. Combined with this, there is scope to further explore the nature of ‘life-deep learning’ for other societal groups, brought by combined religious, moral, ideological and social learning that guides action, beliefs, values, and expression of identity. The LGBT adults in this study demonstrate engagement in distinct forms of life-deep learning to navigate social and moral opprobrium. From this they gain hope, self-respect, empathy with others, and deeper self-knowledge.

Comparative study of human age estimation based on hand-crafted and deep face features

In the past few years, human facial age estimation has drawn a lot of attention in the computer vision and pattern recognition communities because of its important applications in age-based image retrieval, security control and surveillance, biomet- rics, human-computer interaction (HCI) and social robotics. In connection with these investigations, estimating the age of a person from the numerical analysis of his/her face image is a relatively new topic. Also, in problems such as Image Classification the Deep Neural Networks have given the best results in some areas including age estimation. In this work we use three hand-crafted features as well as five deep features that can be obtained from pre-trained deep convolutional neural networks. We do a comparative study of the obtained age estimation results with these features.

Intentional first year curriculum design as a means of facilitating student engagement: some exemplars

The importance of student engagement to higher education quality, making deep learning outcomes possible for students, and achieving student retention, is increasingly being understood. The issue of student engagement in the first year of tertiary study is of particular significance. This paper takes the position that the first year curriculum, and the pedagogical principles that inform its design, are critical influencers of student engagement in the first year learning environment. We use an analysis of case studies prepared for Kift’s ALTC Senior Fellowship to demonstrate ways in which student engagement in the first year of tertiary study can be successfully supported through intentional curriculum design that motivates students to learn, provides a positive learning climate, and encourages students to be active in their learning.

"I'm a designer, get me out of tech class!" : closing the gap between design and technology education

The following paper explores the use of collaborative pedagogical approaches to advance foundational architectural design education, by linking design process to sustainable technology principles. After a brief discussion on architectural design education, the mentioned collaborative approach is described. This approach facilitates students’ exchange of knowledge between two courses, despite no explicit/assessable requirement to do so. The result for the students is deeper learning and a design process that is enriched through collaboration with sustainable technology. The success of this approach has been measured through questionnaires, evaluation surveys, and a comparative assessment of students common to both courses. The paper focuses on the challenges and innovations in connecting architectural design and technology education, where students are encouraged to implement lessons learnt, thereby closing the gap that these courses have traditionally represented.

Experiential ethics education for IT professionals

This chapter presents a new approach to IT ethics education that may be used by teachers in academic institutions, employees responsible for promoting ethics in organisations and individuals wanting to pursue their own professional development. Experiential ethics education emphasises deep learning that prompts a changed experience of ethics. We first consider how this approach complements other ways of engaging in ethics education. We then explore what it means to strive for experiential change and offer a model which may be useful in pursuing IT professional ethics education in this way.

Developing tools to facilitate integrated reflection

Much has been said and documented about the key role that reflection can play in the ongoing development of e-portfolios, particularly e-portfolios utilised for teaching and learning. A review of e-portfolio platforms reveals that a designated space for documenting and collating personal reflections is a typical design feature of both open source and commercial off-the-shelf software. Further investigation of tools within e-portfolio systems for facilitating reflection reveals that, apart from enabling personal journalism through blogs or other writing, scaffolding tools that encourage the actual process of reflection are under-developed. Investigation of a number of prominent e-portfolio projects also reveals that reflection, while presented as critically important, is often viewed as an activity that takes place after a learning activity or experience and not intrinsic to it. This paper assumes an alternative, richer conception of reflection: a process integral to a wide range of activities associated with learning, such as inquiry, communication, editing, analysis and evaluation. Such a conception is consistent with the literature associated with ‘communities of practice’, which is replete with insight into ‘learning through doing’, and with a ‘whole minded’ approach to inquiry. Thus, graduates who are ‘reflective practitioners’ who integrate reflection into their learning will have more to offer a prospective employer than graduates who have adopted an episodic approach to reflection. So, what kinds of tools might facilitate integrated reflection? This paper outlines a number of possibilities for consideration and development. Such tools do not have to be embedded within e-portfolio systems, although there are benefits in doing so. In order to inform future design of e-portfolio systems this paper presents a faceted model of knowledge creation that depicts an ‘ecology of knowing’ in which interaction with, and the production of, learning content is deepened through the construction of well-formed questions of that content. In particular, questions that are initiated by ‘why’ are explored because they are distinguished from the other ‘journalist’ questions (who, what, when, where, and where) in that answers to them demand explanative, as opposed to descriptive, content. They require a rationale. Although why questions do not belong to any one genre and are not simple to classify — responses can contain motivational, conditional, causal, and/or existential content — they do make a difference in the acquisition of understanding. The development of scaffolding that builds on why-questioning to enrich learning is the motivation behind the research that has informed this paper.

Scaffolding for integrated reflection

This paper presents some theoretical perspectives that might inform the design and development of information and communications technology (ICT) tools to support integrated (in-session) reflection and deep learning during e-learning. The role of why questioning provides the focus of discussion and is informed by the literature on critical thinking, sense-making, and reflective practice, as well as recent developments in knowledge management, computational linguistics and automated question generation. It is argued that there exists enormous scope for the development of ICT scaffolding targeted at supporting reflective practice during e-learning. The first generations of e-Portfolio tools provide some evidence for the significance of the benefits of integrating reflection into the design of ICT systems; however, following the review of a number of such systems, as well as a range of ICT applications and services designed to support e-learning, it is argued that the scope of implementation is limited.

Cognitive engagement and questioning online

This chapter discusses a range of issues associated with supporting inquiry and deep reasoning while utilising information and communications technology (ICT). The role of questioning in critical thinking and reflection is considered in the context of scaffolding and new opportunities for ICT-enabled scaffolding identified. In particular, why-questioning provides a key point of focus and is presented as an important consideration in the design of systems that not only require cognitive engagement but aim to nurture it. Advances in automated question generation within intelligent tutoring systems are shown to hold promise for both teaching and learning in a range of other applications. While shortening attention spans appear to be a hazard of engaging with digital media cognitive engagement is presented as something with broader scope than attention span and is best conceived of as a crucible within which a rich mix of cognitive activities take place and from which new knowledge is created.

Responding to the 5Rs : an alternate perspective of Slowmation

This paper is a response to Hoban and Neilsen's (2010) Five Rs model for understanding how learners engage with slowmation. An alternative model (the Learning MMAEPER Model) that builds on the 5Rs model is explained in terms of its use in secondary science preservice teacher education. To probe into the surface and deep learning that can occur during the creation of a slowmation, the learning and relearning model is explored in terms of learning elements. This model can assist teachers to monitor the learning of their students and direct them to a deeper understanding of science concepts.

Whole-of-Curriculum Design in Law

In this book teaching professionalism is characterised by the scholarly underpinning of each contribution; and every contribution provides a rich resource for enhancing teaching practice. The critical concerns for legal education have been identified and discussed: curriculum design that includes graduate attributes; embedding specific attributes across the curriculum; empowering students to learn; academic teamwork to manage large student cohorts; first year and final year transition strategies; tracking students' personal development through the use of ePortfolio; assessment strategies; improving student well-being and promoting resilience; teaching practice to achieve deep learning; flexibility in delivery; the use of Web 2.0 technology; and understanding the 21st century student.

Mathematics, programming, and STEM

Learning mathematics is a complex and dynamic process. In this paper, the authors adopt a semiotic framework (Yeh & Nason, 2004) and highlight programming as one of the main aspects of the semiosis or meaning-making for the learning of mathematics. During a 10-week teaching experiment, mathematical meaning-making was enriched when primary students wrote Logo programs to create 3D virtual worlds. The analysis of results found deep learning in mathematics, as well as in technology and engineering areas. This prompted a rethinking about the nature of learning mathematics and a need to employ and examine a more holistic learning approach for the learning in science, technology, engineering, and mathematics (STEM) areas.

Enhancing human action recognition with region proposals

Deep convolutional network models have dominated recent work in human action recognition as well as image classification. However, these methods are often unduly influenced by the image background, learning and exploiting the presence of cues in typical computer vision datasets. For unbiased robotics applications, the degree of variation and novelty in action backgrounds is far greater than in computer vision datasets. To address this challenge, we propose an “action region proposal” method that, informed by optical flow, extracts image regions likely to contain actions for input into the network both during training and testing. In a range of experiments, we demonstrate that manually segmenting the background is not enough; but through active action region proposals during training and testing, state-of-the-art or better performance can be achieved on individual spatial and temporal video components. Finally, we show by focusing attention through action region proposals, we can further improve upon the existing state-of-the-art in spatio-temporally fused action recognition performance.