Distributed Learning System Research Articles

Managing innovation in educational institutions

Traditionally, project management practices have not drawn any distinction between the characteristics of different projects. Recent research has found that innovative projects require different project management approaches if quality outcomes are to be achieved. Projects involving more innovation require more open management processes and contain a higher percentage of academic and/or professional staff. These characteristics often apply to educational projects in tertiary institutions. There is conflict inherent in managing change projects in educational environments between the classical 'project management' approach and the way that academics and teaching staff, who are independent professionals, traditionally work. This study derived from experience at RMIT, where a system wide online learning system (the Distributed Learning System, DLS) has been implemented since 1999. The implications for the project management approaches suitable for educational projects are discussed and a typology is suggested to categorise projects within an organisation and link them to key factors for successful project management.

Multimedia Distributed Learning Environments: Evolution towards Intelligent Communications

In this paper we describe the realization of a Multimedia Distributed Learning System developed by means of the Java language in a Intranet environment. In particular, we focus our discussion on the definition of the existing scenario, on the requirements specification and on the description of our overall system, by providing also first experimental results. We decided to add to our system a Software Agents based Intelligent Communication Support to manage both high level and low level communication needs coming from the nature of the application. Such communication needs could reasonably be considered as shared by a large family of distributed applications based on Intranet technologies. We present the main characteristics of our agents and describe how to use them to better achieve defined goals in terms of flexibility and scalability and to add network-awareness capabilities to the application.

E-Net: Evolutionary neural network synthesis

E-Net is a new distributed evolutionary learning system that evolves neural-network-based pattern recognition systems (PRSs) with limited human interaction. This system orchestrates a multiplicity of evolutionary and classical learning techniques to synthesize feature detectors, select sets of cooperative features, and assemble classifiers. Feature detectors are represented as feed-forward neural networks and recognition systems are defined using a collection of networks. E-Net evolves network topologies and trains weights to form accurate recognition systems using a computationally efficient process that gradually extends primitive network topologies to form increasingly discriminating structures. The evolutionary search process effectively explores the space of candidate topologies by manipulating populations of feature detectors and recognition systems using variation operators such as crossover and mutation. The majority of evolutionary learning techniques have been designed to perform parameter optimization. E-Net is designed to perform both synthesis and optimization. Consequently, many novel concepts and techniques are introduced in this research that expedite the gradual synthesis of structure, such as the new multitiered selection process used in E-Net's evolutionary algorithm that avoids premature convergence to complex topological structures.

An evaluation of a distributed learning system

Presents an evaluation of a pioneering attempt to deliver a distributed learning Web‐enabled system developed and piloted at RMIT University. Data were gathered from: a survey of students undertaking three undergraduate courses and one postgraduate course at one academic department at RMIT; in‐depth focus group reviews; and lecturer summaries of their experience of using the DLS. One of these courses was offered to international students in Singapore as part of a mixed face‐to‐face and Web‐delivery format. Results indicate that the DLS is generally popular with students. Difficulties were encountered with IT technical support and this provided an unnecessary level of frustration, however, results were generally encouraging. Concludes that the DLS needs to be more effectively used to support a more engaging learning experience and that students need to take more responsibility for their learning.

Distributed Cooperative Bayesian Learning Strategies

This paper addresses the issue of designing an effective distributed learning system in which a number of agent learners estimate the parameter specifying the target probability density in parallel and the population learner (for short, the p-learner) combines their outputs to obtain a significantly better estimate. Such a system is important in speeding up learning. We propose as distributed learning systems two types of the distributed cooperative Bayesian learning strategies (DCB), in which each agent learner or the p-learner employs a probabilistic version of the Gibbs algorithm. We analyze DCBs by giving upper bounds on their average logarithmic losses for predicting probabilities of unseen data as functions of the sample size and the population size. We thereby demonstrate the effectiveness of DCBs by showing that for some probability models, they work approximately (or sometimes exactly) as well as the nondistributed optimal Bayesian strategy, achieving a significant speed-up of learning over it. We also consider the case where the hypothesis class of probability densities is hierarchically parameterized, and there is a feedback of information from the p-learner to agent learners. In this case we propose another type of DCB based on the Markov chain Monte Carlo method, which we abbreviate as HDCB, and characterize its average prediction loss in terms of the number of feedback iterations as well as the population size and the sample size. We thereby demonstrate that for the class of hierarchical Gaussian distributions HDCB works approximately as well as the nondistributed optimal Bayesian strategy, achieving a significant speed-up of learning over it.

Designing distributed learning systems

MANY COMPANIES are adopting an approach of hiring the skills they need, then discarding them and hiring others as needs change. Individuals must rapidly learn in their specialty and also develop new skill sets over the long-term. Access to intellectual capital and to learning and performance support systems is improving with developments in the World Wide Web. However access is not usability, and usability is not organizational impact. This paper provides principles and examples of distributed learning and support systems that can aid individual development. It considers motivation, speed of program development, and cost issues for some current projects where usability and impact are central design criteria.

A Computational Study of Distributed Rule Learning

This report is concerned with a rule learning system called the Distributed Learning System (DLS). Its objective is two-fold: First, as the main contribution, the DLS as a rule-learning technique is described and the resulting computational performance is presented, with definitive computational benefits clearly demonstrated to show the efficacy of using the DLS. Second, the important parameters of the DLS are identified to show the characteristics of the Group Problem Solving (GPS) strategy as implemented in the DLS. On one hand this helps us pinpoint the critical designs of the DLS for effective rule learning; on the other hand this analysis can provide insight into the use of GPS as a more general rule-learning strategy.

DKBLM — Deep knowledge based learning methodology

To solve the Imperfect Theory Problem (ITP) faced by Explanation Based Generalization (EBG), this paper proposes a methodology, Deep Knowledge Based Learning Methodology (DKBLM) by name, and gives an implementation of DKBLM, called Hierarchically Distributed Learning System (HDLS). As an example of HDLS's application, this paper shows a learning system (MLS) in meteorology domain and its running with a simplified example. DKBLM can acquire experiential knowledge with causality in it. It is applicable to those kinds of domains, in which experiments are relatively difficult to carry out, and in which there exist many available knowledge systems at different levels for the same domain (such as weather forecasting).

Learning control strategies for chemical processes: a distributed approach

The design of a distributed learning system (DLS) which combines the features of instance-space and hypothesis-space methods is described. This algorithm decomposes a data set of training examples into subsets. After applying an inductive learning program on each subset, it synthesizes the results using a genetic algorithm. It is shown that this parallel distributed approach is more efficient, since each inductive learning program works on only a subset of data. Since the genetic algorithm searches globally in the hypothesis space, this approach gives a more accurate concept description. The implementation of DLS in Common LISP is discussed, and its distributed approach is compared to C4.5 and PLS1 algorithms.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>