Learning analytics beyond the LMS : the Connected Learning Analytics toolkit

We present a Connected Learning Analytics (CLA) toolkit, which enables data to be extracted from social media and imported into a Learning Record Store (LRS), as defined by the new xAPI standard. Core to the toolkit is the notion of learner access to their own data. A number of implementational issues are discussed, and an ontology of xAPI verb/object/activity statements as they might be unified across 7 different social media and online environments is introduced. After considering some of the analytics that learners might be interested in discovering about their own processes (the delivery of which is prioritised for the toolkit) we propose a set of learning activities that could be easily implemented, and their data tracked by anyone using the toolkit and a LRS.

Improving robot vision models for object detection through interaction

We propose a method for learning specific object representations that can be applied (and reused) in visual detection and identification tasks. A machine learning technique called Cartesian Genetic Programming (CGP) is used to create these models based on a series of images. Our research investigates how manipulation actions might allow for the development of better visual models and therefore better robot vision. This paper describes how visual object representations can be learned and improved by performing object manipulation actions, such as, poke, push and pick-up with a humanoid robot. The improvement can be measured and allows for the robot to select and perform the `right' action, i.e. the action with the best possible improvement of the detector.

A perfusion fMRI investigation of thematic and categorical context effects in the spoken production of object names

The context in which objects are presented influences the speed at which they are named. We employed the blocked cyclic naming paradigm and perfusion functional magnetic resonance imaging (fMRI) to investigate the mechanisms responsible for interference effects reported for thematicallyand categorically related compared to unrelated contexts. Naming objects in categorically homogeneous contexts induced a significant interference effect that accumulated from the second cycle onwards. This interference effect was associated with significant perfusion signal decreases in left middle and posterior lateral temporal cortex and the hippocampus. By contrast, thematically homogeneous contexts facilitated naming latencies significantly in the first cycle and did not differ from heterogeneous contexts thereafter, nor were they associated with any perfusion signal changes compared to heterogeneous contexts. These results are interpreted as being consistent with an account in which the interference effect both originates and has its locus at the lexical level, with an incremental learning mechanism adapting the activation levels of target lexical representations following access. We discuss the implications of these findings for accounts that assume thematic relations can be active lexical competitors or assume mandatory involvement of top-down control mechanisms in interference effects during naming.

Improving deep convultional neural networks with unsupervised feature learning

The latest generation of Deep Convolutional Neural Networks (DCNN) have dramatically advanced challenging computer vision tasks, especially in object detection and object classification, achieving state-of-the-art performance in several computer vision tasks including text recognition, sign recognition, face recognition and scene understanding. The depth of these supervised networks has enabled learning deeper and hierarchical representation of features. In parallel, unsupervised deep learning such as Convolutional Deep Belief Network (CDBN) has also achieved state-of-the-art in many computer vision tasks. However, there is very limited research on jointly exploiting the strength of these two approaches. In this paper, we investigate the learning capability of both methods. We compare the output of individual layers and show that many learnt filters and outputs of the corresponding level layer are almost similar for both approaches. Stacking the DCNN on top of unsupervised layers or replacing layers in the DCNN with the corresponding learnt layers in the CDBN can improve the recognition/classification accuracy and training computational expense. We demonstrate the validity of the proposal on ImageNet dataset.

The Networks of Learning in Technological Innovation : The Emergence of Collaboration Across Fields of Expertise

This study addresses four issues concerning technological product innovations. First, the nature of the very early phases or "embryonic stages" of technological innovation is addressed. Second, this study analyzes why and by what means people initiate innovation processes outside the technological community and the field of expertise of the established industry. In other words, this study addresses the initiation of innovation that occurs without the expertise of established organizations, such as technology firms, professional societies and research institutes operating in the technological field under consideration. Third, the significance of interorganizational learning processes for technological innovation is dealt with. Fourth, this consideration is supplemented by considering how network collaboration and learning change when formalized product development work and the commercialization of innovation advance. These issues are addressed through the empirical analysis of the following three product innovations: Benecol margarine, the Nordic Mobile Telephone system (NMT) and the ProWellness Diabetes Management System (PDMS). This study utilizes the theoretical insights of cultural-historical activity theory on the development of human activities and learning. Activity-theoretical conceptualizations are used in the critical assessment and advancement of the concept of networks of learning. This concept was originally proposed by the research group of organizational scientist Walter Powell. A network of learning refers to the interorganizational collaboration that pools resources, ideas and know-how without market-based or hierarchical relations. The concept of an activity system is used in defining the nodes of the networks of learning. Network collaboration and learning are analyzed with regard to the shared object of development work. According to this study, enduring dilemmas and tensions in activity explain the participants' motives for carrying out actions that lead to novel product concepts in the early phases of technological innovation. These actions comprise the initiation of development work outside the relevant fields of expertise and collaboration and learning across fields of expertise in the absence of market-based or hierarchical relations. These networks of learning are fragile and impermanent. This study suggests that the significance of networks of learning across fields of expertise becomes more and more crucial for innovation activities.

Learning challenges of NGOs in development : Co-operation of Finnish NGOs in Morogoro, Tanzania

Non-governmental organisations (NGOs) have gained an important role in development co-operation during the last two decades. The development funding channelled through NGOs has increased and the number of NGOs engaged in development activities, both North and South, has been growing. Supporting NGOs has been seen as one way to strengthen civil society in the South and to provide potential for enhancing more effective development than the state, and to exercise participatory development and partnership in their North-South relationships. This study focuses on learning in the co-operation practices of small Finnish NGOs in Morogoro, Tanzania. Drawing on the cultural-historical activity theory and the theory of expansive learning, in this study I understand learning as a qualitative change in the actual co-operation practices. The qualitative change, for its part, emerges out of attempts to deal with the contradictions in the present activity. I use the concept of developmental contradiction in exploring the co-operation of the small Finnish NGOs with their Tanzanian counterparts. Developmental contradiction connects learning to actual practice and its historical development. By history, in this study I refer to multiple developmental trajectories, such as trajectories of individual participants, organisations, co-operation practices and the institutional system in which the NGO-development co-operation is embedded. In the empirical chapters I explore the co-operation both in the development co-operation projects and in micro-level interaction between partners taking place within the projects. I analyse the perceptions of the Finnish participants about the different developmental trajectories, the tensions, inclusions and exclusions in the evolving object of co-operation in one project, the construction of power relations in project meetings in three projects, and the collision of explicated partnership with the emerging practice of trusteeship in one project. On the basis of the empirical analyses I elaborate four developmental contradictions and learning challenges for the co-operation. The developmental contradictions include: 1) implementing a ready-made Finnish project idea vs. taking the current activities of Tanzanian NGO as a starting point; 2) gaining experiences and cultural interaction vs. access to outside funding; 3) promoting the official tools of development co-operation in training vs. use of tools and procedures taken from the prior activities of both partners in actual practice; and 4) asymmetric relations between the partners vs. rhetoric of equal partnership. Consequently, on the basis of developmental contradictions four learning challenges are suggested: a shift from legitimation of Finnish ideas to negotiation, transcending the separate objects and finding a partly joint object, developing locally shared tools for the co-operation, and identification and reflection of the power relations in the practice of co-operation. Keywords: activity theory; expansive learning; NGO development co-operation; partnership; power

Learning Nonlinear Visual Processing from Natural Images

The paradigm of computational vision hypothesizes that any visual function -- such as the recognition of your grandparent -- can be replicated by computational processing of the visual input. What are these computations that the brain performs? What should or could they be? Working on the latter question, this dissertation takes the statistical approach, where the suitable computations are attempted to be learned from the natural visual data itself. In particular, we empirically study the computational processing that emerges from the statistical properties of the visual world and the constraints and objectives specified for the learning process. This thesis consists of an introduction and 7 peer-reviewed publications, where the purpose of the introduction is to illustrate the area of study to a reader who is not familiar with computational vision research. In the scope of the introduction, we will briefly overview the primary challenges to visual processing, as well as recall some of the current opinions on visual processing in the early visual systems of animals. Next, we describe the methodology we have used in our research, and discuss the presented results. We have included some additional remarks, speculations and conclusions to this discussion that were not featured in the original publications. We present the following results in the publications of this thesis. First, we empirically demonstrate that luminance and contrast are strongly dependent in natural images, contradicting previous theories suggesting that luminance and contrast were processed separately in natural systems due to their independence in the visual data. Second, we show that simple cell -like receptive fields of the primary visual cortex can be learned in the nonlinear contrast domain by maximization of independence. Further, we provide first-time reports of the emergence of conjunctive (corner-detecting) and subtractive (opponent orientation) processing due to nonlinear projection pursuit with simple objective functions related to sparseness and response energy optimization. Then, we show that attempting to extract independent components of nonlinear histogram statistics of a biologically plausible representation leads to projection directions that appear to differentiate between visual contexts. Such processing might be applicable for priming, \ie the selection and tuning of later visual processing. We continue by showing that a different kind of thresholded low-frequency priming can be learned and used to make object detection faster with little loss in accuracy. Finally, we show that in a computational object detection setting, nonlinearly gain-controlled visual features of medium complexity can be acquired sequentially as images are encountered and discarded. We present two online algorithms to perform this feature selection, and propose the idea that for artificial systems, some processing mechanisms could be selectable from the environment without optimizing the mechanisms themselves. In summary, this thesis explores learning visual processing on several levels. The learning can be understood as interplay of input data, model structures, learning objectives, and estimation algorithms. The presented work adds to the growing body of evidence showing that statistical methods can be used to acquire intuitively meaningful visual processing mechanisms. The work also presents some predictions and ideas regarding biological visual processing.

Discriminative learning of Bayesian networks via factorized conditional log-likelihood

We propose an efficient and parameter-free scoring criterion, the factorized conditional log-likelihood (ˆfCLL), for learning Bayesian network classifiers. The proposed score is an approximation of the conditional log-likelihood criterion. The approximation is devised in order to guarantee decomposability over the network structure, as well as efficient estimation of the optimal parameters, achieving the same time and space complexity as the traditional log-likelihood scoring criterion. The resulting criterion has an information-theoretic interpretation based on interaction information, which exhibits its discriminative nature. To evaluate the performance of the proposed criterion, we present an empirical comparison with state-of-the-art classifiers. Results on a large suite of benchmark data sets from the UCI repository show that ˆfCLL-trained classifiers achieve at least as good accuracy as the best compared classifiers, using significantly less computational resources.

Variable Sparsity Kernel Learning

This paper(1) presents novel algorithms and applications for a particular class of mixed-norm regularization based Multiple Kernel Learning (MKL) formulations. The formulations assume that the given kernels are grouped and employ l(1) norm regularization for promoting sparsity within RKHS norms of each group and l(s), s >= 2 norm regularization for promoting non-sparse combinations across groups. Various sparsity levels in combining the kernels can be achieved by varying the grouping of kernels-hence we name the formulations as Variable Sparsity Kernel Learning (VSKL) formulations. While previous attempts have a non-convex formulation, here we present a convex formulation which admits efficient Mirror-Descent (MD) based solving techniques. The proposed MD based algorithm optimizes over product of simplices and has a computational complexity of O (m(2)n(tot) log n(max)/epsilon(2)) where m is no. training data points, n(max), n(tot) are the maximum no. kernels in any group, total no. kernels respectively and epsilon is the error in approximating the objective. A detailed proof of convergence of the algorithm is also presented. Experimental results show that the VSKL formulations are well-suited for multi-modal learning tasks like object categorization. Results also show that the MD based algorithm outperforms state-of-the-art MKL solvers in terms of computational efficiency.

Experiments on three-dimensional wire-frame object recognition

Some experimental results on the recognition of three-dimensional wire-frame objects are presented. In order to overcome the limitations of a recent model, which employs radial basis functions-based neural networks, we have proposed a hybrid learning system for object recognition, featuring: an optimization strategy (simulated annealing) in order to avoid local minima of an energy functional; and an appropriate choice of centers of the units. Further, in an attempt to achieve improved generalization ability, and to reduce the time for training, we invoke the principle of self-organization which utilises an unsupervised learning algorithm.

Priestley College: How do you successfully develop your new learning space? ...ask your students, of course!

When Priestley College began to plan the redevelopment of its learning resource centre, it continued the culture of student involvement that exists within the College by asking students to help plan and create the new development. This case study describes how the Jisc infoKit on 'Planning and Designing Technology-Rich Learning Spaces' was used as the starting point for ideas and planning, and how the finished development was the recognisable result of students' ideas and plans.

Xaverian College: Seeing the long term benefit of personalised learning with bespoke VLE

Xaverian College has created a bespoke web-based solution for the academic, pastoral, managerial and administrative requirements of the College that is accessible to all staff and students. Their VLE/ILP intranet system is cutting edge as it offers a remarkable degree of integration because of its holistic approach and is very ‘student-friendly’.

Essex Adult Community Learning: PTLLS for distance and blended learning

Funded by the LSIS Excellence and Improvement Fund, Essex Adult Community Learning has produced a distance/blended learning resource on CD-Rom for tutors in the post-16 sector to achieve the City & Guilds PTLLS (7303 Preparing to Teach in the Lifelong Learning Sector) at Level 4. The aim of the resource is to improve teaching and learning by providing a flexible way to access initial teacher training where candidates may otherwise find it difficult or impossible to attend a taught course. It is also intended to increase tutors' own e-learning skills.

People and Business Development (PBD): Learning to deliver learning effectively

In October 2010, PBD introduced its eQual Learning VLE (virtual learning environment) to provide an online knowledge resource for its students. During the project, the company learnt many lessons about how to deliver learning effectively. In the course of a year researching VLE platforms, looking for material, and remapping NVQ courses for new QCF qualifications, the company realised that it was more important to deliver engaging and relevant content, rather than boasting the most innovative technology.