Thesis (M.Sc.). / Written for the School of Computer Science. Title from title page of PDF (viewed 2009/06/18). Includes bibliographical references.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1969. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references.

Thesis (Ph. D.)--University of British Columbia, 2000. / Description based on contents viewed Aug. 16, 2007; title from title screen. Includes bibliographical references (p. 153-157).

Developments in scanning transmission electron microscopy (STEM) have opened up new possibilities for time-resolved imaging at the atomic scale. However, rapid imaging of single atom dynamics brings with it a new set of challenges, particularly regarding noise and the interaction between the electron beam and the specimen. This thesis develops a set of analytical tools for capturing atomic motion and analyzing the dynamic behaviour of materials at the atomic scale. Machine learning is increasingly playing an important role in the analysis of electron microscopy data. In this light, new unsupervised learning tools are developed here for noise removal under low-dose imaging conditions and for identifying the motion of surface atoms. The scope for real-time processing and analysis is also explored, which is of rising importance as electron microscopy datasets grow in size and complexity. These advances in image processing and analysis are combined with computational modelling to uncover new chemical and physical insights into the motion of atoms adsorbed onto surfaces. Of particular interest are systems for heterogeneous catalysis, where the catalytic activity can depend intimately on the atomic environment. The study of Cu atoms on a graphene oxide support reveals that the atoms undergo anomalous diffusion as a result of spatial and energetic disorder present in the substrate. The investigation is extended to examine the structure and stability of small Cu clusters on graphene oxide, with atomistic modelling used to understand the significant role played by the substrate. Finally, the analytical methods are used to study the surface reconstruction of silicon alongside the electron beam-induced motion of adatoms on the surface. Taken together, these studies demonstrate the materials insights that can be obtained with time-resolved STEM imaging, and highlight the importance of combining state-ofthe- art imaging with computational analysis and atomistic modelling to quantitatively characterize the behaviour of materials with atomic resolution.

Most real-world data such as images and videos are naturally organized as tensors, and often have high dimensionality. Tensor subspace learning is a fundamental problem that aims at finding low-dimensional representations from tensors while preserving their intrinsic characteristics. By dealing with tensors in the learned subspace, subsequent tasks such as clustering, classification, visualization, and interpretation can be greatly facilitated. This thesis studies the tensor subspace learning problem from a generative perspective, and proposes four probabilistic methods that generalize the ideas of classical subspace learning techniques for tensor analysis. Probabilistic Rank-One Tensor Analysis (PROTA) generalizes probabilistic principle component analysis. It is flexible in capturing data characteristics, and avoids rotational ambiguity. For robustness against overfitting, concurrent regularizations are further proposed to concurrently and coherently penalize the whole subspace, so that unnecessary scale restrictions can be relaxed in regularizing PROTA. Probabilistic Rank-One Discriminant Analysis (PRODA) is a bilinear generalization of probabilistic linear discriminant analysis. It learns a discriminative subspace by representing each observation as a linear combination of collective and individual rank-one matrices. This provides PRODA with both the expressiveness of capturing discriminative features and non-discriminative noise, and the capability of exploiting the (2D) tensor structures. Bilinear Probabilistic Canonical Correlation Analysis (BPCCA) generalizes probabilistic canonical correlation analysis for learning correlations between two sets of matrices. It is built on a hybrid Tucker model in which the two-view matrices are combined in two stages via matrix-based and vector-based concatenations, respectively. This enables BPCCA to capture two-view correlations without breaking the matrix structures. Bayesian Low-Tubal-Rank Tensor Factorization (BTRTF) is a fully Bayesian treatment of robust principle component analysis for recovering tensors corrupted with gross outliers. It is based on the recently proposed tensor-SVD model, and has more expressive modeling power in characterizing tensors with certain orientation such as images and videos. A novel sparsity-inducing prior is also proposed to provide BTRTF with automatic determination of the tensor rank (subspace dimensionality). Comprehensive validations and evaluations are carried out on both synthetic and real-world datasets. Empirical studies on parameter sensitivities and convergence properties are also provided. Experimental results show that the proposed methods achieve the best overall performance in various applications such as face recognition, photograph-sketch match, and background modeling. Keywords: Tensor subspace learning, probabilistic models, Bayesian inference, tensor decomposition.

Submitted to the Faculty of Education in fulfilment of the requirements for the degree of MASTER OF SCIENCE EDUCATION in the Department of Mathematics, Science and Technology (MSTE) at the University of Zululand, 2015. / This study set out to investigate beliefs held by pre service Bachelor of Education (B.Ed) students about the nature of science and science teaching and learning. The research sample comprised one hundred and eighty four (184) third and fourth year (B.Ed) students majoring in mathematics and physical sciences. Data on students’ epistemological beliefs about the nature of science and science teaching and Learning were collected using two questionnaires: The Nature of Science as Argument Questionnaire (NSAAQ) and Beliefs About Reformed Science Teaching and Learning (BARSTL). Furthermore the study sought to find out if those beliefs cohered with the beliefs espoused by the National Curriculum Statement (NCS) for Physical Sciences grades 10-12. The conceptual framework of this study was framed upon the preponderance of literature that carried the view that a teacher’s classroom practices are a consequence of two main dialectic influences: (a) the teacher’s epistemological beliefs about the nature of science, which may be either naïve or sophisticated; and (b) the teacher’s beliefs about teaching and learning, which may be either traditional or reformed. Accordingly, the conceptual framework guiding the study opined that teachers holding naïve beliefs about the nature of science, and those holding traditional notions of teaching and learning will be characterized by teacher-centred instructional approaches, while those holding sophisticated beliefs of the nature of science and a reformed view of teaching and learning will be associated with learner-centred instructional approaches. This study was a case study conducted at a South African university, and involved one hundred and eighty-four third and fourth year students registered for a four-year Bachelor of Education (B.Ed) degree for the Senior and Further Education and Training phase. During these two final years of the programme students are engaged in science enquiry practices in their Methods modules. The participants were registered in physical science and mathematics education. Intact groups were used, so there was no sampling undertaken to select participants. Data were collected by the use of (a) the Nature of Science as Argument Questionnaire (NSAAQ), to determine epistemological beliefs held by the participants about the nature of science, as well as the concurrence of those beliefs with the views about science teaching and learning espoused by the NCS; and (b) the Beliefs about Reformed Science Teaching and Learning (BARSTL) questionnaire, to determine the beliefs held by preservice education students about science teaching and learning. Data analysis involved the use of both descriptive statistical methods to decipher patterns and general trends regarding the epistemological beliefs about science held by participants, and their beliefs about science teaching and learning, as well as inferential statistics to test both a priori and a posteriori hypotheses. Similarly, statistical analysis was carried out to determine whether or not third- and fourth-year pre-service science education students held beliefs about science teaching and learning that were in agreement with the pedagogical content beliefs about science teaching and learning espoused by the NCS. The study found that pre service students held significantly more sophisticated epistemological beliefs about the nature of science at fourth year than at third year level. The results also showed that fourth year students demonstrated a significantly higher level of ‘reformed oriented teaching and learning beliefs’ about science than did the third year students. The results however showed that third and fourth year students held beliefs that were not in line with the beliefs espoused by the National Curriculum Statement (NCS). These results support studies which have found that student teachers become more sophisticated in their epistemological beliefs towards graduation. The findings also showed that the B.Ed programme is succeeding in developing both epistemological beliefs about the nature of science and teaching and learning. The degree to which the programme succeeded in developing these beliefs was however quite small. This study recommends that further investigations be done to determine whether students who hold sophisticated epistemological beliefs about the nature of science and ‘reformed beliefs about science teaching and learning’ also demonstrate superior science teaching skills

This thesis explores the use of invariant features for learning sprites from image sequences, and for recognising object categories in images. A popular framework for the interpretation of image sequences is the layers or sprite model of e.g. Wang and Adelson (1994), Irani et al. (1994). Jojic and Frey (2001) provide a generative probabilistic model framework for this task, but their algorithm is slow as it needs to search over discretised transformations (e.g. translations, or affines) for each layer. We show that by using invariant features (e.g. Lowe’s SIFT features) and clustering their motions we can reduce or eliminate the search and thus learn the sprites much faster. The algorithm is demonstrated on example image sequences. We introduce the Generative Template of Features (GTF), a parts-based model for visual object category detection. The GTF consists of a number of parts, and for each part there is a corresponding spatial location distribution and a distribution over ‘visual words’ (clusters of invariant features). We evaluate the performance of the GTF model for object localisation as compared to other techniques, and show that such a relatively simple model can give state-of- the-art performance. We also discuss the connection of the GTF to Hough-transform-like methods for object localisation.

A key goal for humans and artificial intelligence systems is to develop an accurate and unified picture of the outside world based on the data from any sense(s) that may be available. The availability of multiple senses presents the perceptual system with new opportunities to fulfil this goal, but exploiting these opportunities first requires the solution of two related tasks. The first is how to make the best use of any redundant information from the sensors to produce the most accurate percept of the state of the world. The second is how to interpret the relationship between observations in each modality; for example, the correspondence problem of whether or not they originate from the same source. This thesis investigates these questions using ideal Bayesian observers as the underlying theoretical approach. In particular, the latter correspondence task is treated as a problem of Bayesian model selection or structure inference in Bayesian networks. This approach provides a unified and principled way of representing and understanding the perceptual problems faced by humans and machines and their commonality. In the domain of machine intelligence, we exploit the developed theory for practical benefit, developing a model to represent audio-visual correlations. Unsupervised learning in this model provides automatic calibration and user appearance learning, without human intervention. Inference in the model involves explicit reasoning about the association between latent sources and observations. This provides audio-visual tracking through occlusion with improved accuracy compared to standard techniques. It also provides detection, verification and speech segmentation, ultimately allowing the machine to understand ``who said what, where?'' in multi-party conversations. In the domain of human neuroscience, we show how a variety of recent results in multimodal perception can be understood as the consequence of probabilistic reasoning about the causal structure of multimodal observations. We show this for a localisation task in audio-visual psychophysics, which is very similar to the task solved by our machine learning system. We also use the same theory to understand results from experiments in the completely different paradigm of oddity detection using visual and haptic modalities. These results begin to suggest that the human perceptual system performs -- or at least approximates -- sophisticated probabilistic reasoning about the causal structure of observations under the hood.

e-Research in the Life Sciences examines the use of online tools in the life sciences and finds that their use has significant impact, namely the formation of a Scientific/Intellectual Movement (SIM) (Frickel & Gross, 2005) complemented by a Computerisation Movement (CM) (Kling & Iacono, 1994) which is mobilising global electronic resources to form visible colleges of life science researchers, who are enrolling others and successfully promoting their open science goals via mainstream scientific literature. Those within this movement are also using these online tools to change their work practices, producing scientific knowledge in a highly networked and distributed group which has less regard for traditional institutional and disciplinary boundaries. This thesis, by combining ideas about SIMs and CMs, fills a gap in research that is typically confined to treating new tools as a part of scientific communication or in specialist areas like distributed collaboration but not in terms of broader changes in science. Case studies have been conducted for three types of online tools: the scientific social networking tool FriendFeed, open laboratory notebooks, and science blogs. Data have been collected from semi-structured interviews, and the online writings of research participants. The case studies of exemplary use by scientists of the web form a baseline for future studies in the area. Boundaries between formal and informal scholarly communication are now blurred. At the formal level, which peer-reviewed print journals continue, many academic publishers now also have online open access, frequently in advance of print publication. At the informal level, what used to be confined to water-cooler chat and the conference circuit is now also discussed on mailing lists, forums and blogs (Borgman, 2007). As these online tools generate new practices they have potential to affect future academic assessment and dissemination practices.

Based on the idea of broadening access to learning opportunities for all by means of new Information and Communication Technologies (ICTs), the Open Educational Resources (OER) movement has gained ground during the first years of the 21st Century while capturing the imagination of educators, university leaders, policy-makers and opinion leaders all over the globe. Drawing on socio-technical theories and adopting a case study research design, which involved the analysis of both qualitative and quantitative data, this thesis addresses the manifold tensions and paradoxes that may emerge out of the interplay between a highly predefined model of OER provision and the everyday realities and institutional contexts of different higher education settings. In particular, it focuses on the process of implementation by Spanish universities of OpenCourseWare (OCW) initiatives, a widely adopted model of OER provision that was originally devised at the Massachusetts Institute of Technology. By examining the enactment of technology as a situated phenomenon, this study sheds light on the roles that OCW initiatives play in relation to the strategic orientation of universities and how the actual involvement of scholars in the creation of this type of materials is often curbed by some entrenched institutional arrangements and prevailing academic cultures. The findings of this thesis have theoretical as well as practical implications, which suggest that the replication of models of OER provision outside the specific settings in which they were originally devised is a rather problematic endeavour. More generally, it supports the idea that the implementation of ICTs must be always accompanied by social structures that are mindful and respectful of local specificities and institutional arrangements. Another key conclusion is that, if universities are genuinely committed to broadening access to higher education opportunities and supporting participation in life-long learning by means of ICTs, it is crucial to understand the ways and extent to which OER initiatives can actually contribute to achieving such goals.