Improved SOM learning using simulated annealing

Self-Organizing Map (SOM) algorithm has been extensively used for analysis and classification problems. For this kind of problems, datasets become more and more large and it is necessary to speed up the SOM learning. In this paper we present an application of the Simulated Annealing (SA) procedure to the SOM learning algorithm. The goal of the algorithm is to obtain fast learning and better performance in terms of matching of input data and regularity of the obtained map. An advantage of the proposed technique is that it preserves the simplicity of the basic algorithm. Several tests, carried out on different large datasets, demonstrate the effectiveness of the proposed algorithm in comparison with the original SOM and with some of its modification introduced to speed-up the learning.

The origin, molecular regulation and therapeutic potential of myogenic stem cell populations

Satellite cells, originating in the embryonic dermamyotome, reside beneath the myofibre of mature adult skeletal muscle and constitute the tissue-specific stem cell population. Recent advances following the identification of markers for these cells (including Pax7, Myf5, c-Met and CD34) (CD, cluster of differentiation; c-Met, mesenchymal epithelial transition factor) have led to a greater understanding of the role played by satellite cells in the regeneration of new skeletal muscle during growth and following injury. In response to muscle damage, satellite cells harbour the ability both to form myogenic precursors and to self-renew to repopulate the stem cell niche following myofibre damage. More recently, other stem cell populations including bone marrow stem cells, skeletal muscle side population cells and mesoangioblasts have also been shown to have myogenic potential in culture, and to be able to form skeletal muscle myofibres in vivo and engraft into the satellite cell niche. These cell types, along with satellite cells, have shown potential when used as a therapy for skeletal muscle wasting disorders where the intrinsic stem cell population is genetically unable to repair non-functioning muscle tissue. Accurate understanding of the mechanisms controlling satellite cell lineage progression and self-renewal as well as the recruitment of other stem cell types towards the myogenic lineage is crucial if we are to exploit the power of these cells in combating myopathic conditions. Here we highlight the origin, molecular regulation and therapeutic potential of all the major cell types capable of undergoing myogenic differentiation and discuss their potential therapeutic application.

A study of students' metacognitive beliefs about foreign language study and their impact on learning

This article reports on an investigation into the language learning beliefs of students of French in England, aged 16 to 18. It focuses on qualitative data from two groups of learners (10 in total). While both groups had broadly similar levels of achievement in French in terns of examination success, they dffered greatly in the self-image they had of themselves as language learners, with one group displaying low levels of self-eficacy beliefs regarding the possibility of future success. The implica tions of such beliefs for students' levels of motivation and persistence are discussed, together with their possible causes. The article concludes by suggesting changes in classroom practice that might help students develop a more positive image of them selves as language learners.

Learner strategies and self-efficacy: making the connection

This article reports on part of a larger study of the impact of strategy training in listening on learners of French, aged 16 to 17. One aim of the project was to investigate whether such training might have a positive effect on the self-efficacy of learners, by helping them see the relationship between the strategies they employed and what they achieved. One group of learners, as well as receiving strategy training, also received detailed feedback on their listening strategy use and on the reflective diaries they were asked to keep, in order to draw their attention to the relationship between strategies and learning outcomes. Another group received strategy training without feedback or reflective diaries, while a comparison group received neither strategy training nor feedback. As a result of the training, there was some evidence that students who had received feedback had made the biggest gains in certain aspects of self-efficacy for listening; although their gains as compared to the non-feedback group were not as great as had been anticipated. Reasons for this are discussed. The article concludes by suggesting changes in how teachers approach listening comprehension that may improve learners' view of themselves as listeners.

Giving up on modern foreign languages? Students' perceptions of learning French

This article reports on the findings of an investigation into the attitudes of English students aged 16 to 19 years towards French and how they view the reasons behind their level of achievement. Those students who attributed success to effort, high ability, and effective learning strategies had higher levels of achievement, and students intending to continue French after age 16 were more likely than noncontinuers to attribute success to these factors. Low ability and task difficulty were the main reasons cited for lack of achievement in French, whereas the possible role of learning strategies tended to be overlooked by students. It is argued that learners' self-concept and motivation might be enhanced through approaches that encourage learners to explore the causal links between the strategies they employ and their academic performance, thereby changing the attributions they make for success or failure.

Regulation of the neural circuitry of emotion by compassion meditation: effects of meditative expertise

Recent brain imaging studies using functional magnetic resonance imaging (fMRI) have implicated insula and anterior cingulate cortices in the empathic response to another's pain. However, virtually nothing is known about the impact of the voluntary generation of compassion on this network. To investigate these questions we assessed brain activity using fMRI while novice and expert meditation practitioners generated a loving-kindness-compassion meditation state. To probe affective reactivity, we presented emotional and neutral sounds during the meditation and comparison periods. Our main hypothesis was that the concern for others cultivated during this form of meditation enhances affective processing, in particular in response to sounds of distress, and that this response to emotional sounds is modulated by the degree of meditation training. The presentation of the emotional sounds was associated with increased pupil diameter and activation of limbic regions (insula and cingulate cortices) during meditation (versus rest). During meditation, activation in insula was greater during presentation of negative sounds than positive or neutral sounds in expert than it was in novice meditators. The strength of activation in insula was also associated with self-reported intensity of the meditation for both groups. These results support the role of the limbic circuitry in emotion sharing. The comparison between meditation vs. rest states between experts and novices also showed increased activation in amygdala, right temporo-parietal junction (TPJ), and right posterior superior temporal sulcus (pSTS) in response to all sounds, suggesting, greater detection of the emotional sounds, and enhanced mentation in response to emotional human vocalizations for experts than novices during meditation. Together these data indicate that the mental expertise to cultivate positive emotion alters the activation of circuitries previously linked to empathy and theory of mind in response to emotional stimuli.

Example uses of a learning object

A self study course for learning to program using the C programming language has been developed. A Learning Object approach was used in the design of the course. One of the benefits of the Learning Object approach is that the learning material can be reused for different purposes. 'Me course developed is designed so that learners can choose the pedagogical approach most suited to their personal learning requirements. For all learning approaches a set of common Assessment Learning Objects (ALOs or tests) have been created. The design of formative assessments with ALOs can be carried out by the Instructional Designer grouping ALOs to correspond to a specific assessment intention. The course is non-credit earning, so there is no summative assessment, all assessment is formative. In this paper examples of ALOs and their uses is presented together with their uses as decided by the Instructional Designer and learner. Personalisation of the formative assessment of skills can be decided by the Instructional Designer or the learner using a repository of pre-designed ALOs. The process of combining ALOs can be carried out manually or in a semi-automated way using metadata that describes the ALO and the skill it is designed to assess.

Clustering quality and topology preservation in fast learning SOMs

The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for data represented in multidimensional input spaces. In this paper, we describe Fast Learning SOM (FLSOM) which adopts a learning algorithm that improves the performance of the standard SOM with respect to the convergence time in the training phase. We show that FLSOM also improves the quality of the map by providing better clustering quality and topology preservation of multidimensional input data. Several tests have been carried out on different multidimensional datasets, which demonstrate better performances of the algorithm in comparison with the original SOM.

A neural network enhanced generalized minimum variance self-tuning proportional, integral and derivative control algorithm for complex dynamic systems

A neural network enhanced proportional, integral and derivative (PID) controller is presented that combines the attributes of neural network learning with a generalized minimum-variance self-tuning control (STC) strategy. The neuro PID controller is structured with plant model identification and PID parameter tuning. The plants to be controlled are approximated by an equivalent model composed of a simple linear submodel to approximate plant dynamics around operating points, plus an error agent to accommodate the errors induced by linear submodel inaccuracy due to non-linearities and other complexities. A generalized recursive least-squares algorithm is used to identify the linear submodel, and a layered neural network is used to detect the error agent in which the weights are updated on the basis of the error between the plant output and the output from the linear submodel. The procedure for controller design is based on the equivalent model, and therefore the error agent is naturally functioned within the control law. In this way the controller can deal not only with a wide range of linear dynamic plants but also with those complex plants characterized by severe non-linearity, uncertainties and non-minimum phase behaviours. Two simulation studies are provided to demonstrate the effectiveness of the controller design procedure.

Self-organising neural networks for adaptive control

Self-organizing neural networks have been implemented in a wide range of application areas such as speech processing, image processing, optimization and robotics. Recent variations to the basic model proposed by the authors enable it to order state space using a subset of the input vector and to apply a local adaptation procedure that does not rely on a predefined test duration limit. Both these variations have been incorporated into a new feature map architecture that forms an integral part of an Hybrid Learning System (HLS) based on a genetic-based classifier system. Problems are represented within HLS as objects characterized by environmental features. Objects controlled by the system have preset targets set against a subset of their features. The system's objective is to achieve these targets by evolving a behavioural repertoire that efficiently explores and exploits the problem environment. Feature maps encode two types of knowledge within HLS — long-term memory traces of useful regularities within the environment and the classifier performance data calibrated against an object's feature states and targets. Self-organization of these networks constitutes non-genetic-based (experience-driven) learning within HLS. This paper presents a description of the HLS architecture and an analysis of the modified feature map implementing associative memory. Initial results are presented that demonstrate the behaviour of the system on a simple control task.

Using self-organising feature maps for the control of artificial organisms

Variations on the standard Kohonen feature map can enable an ordering of the map state space by using only a limited subset of the complete input vector. Also it is possible to employ merely a local adaptation procedure to order the map, rather than having to rely on global variables and objectives. Such variations have been included as part of a hybrid learning system (HLS) which has arisen out of a genetic-based classifier system. In the paper a description of the modified feature map is given, which constitutes the HLSs long term memory, and results in the control of a simple maze running task are presented, thereby demonstrating the value of goal related feedback within the overall network.

Amygdalar function reflects common individual differences in emotion and pain regulation success

Although the co-occurrence of negative affect and pain is well recognized, the mechanism underlying their association is unclear. To examine whether a common self-regulatory ability impacts the experience of both emotion and pain, we integrated neuroimaging, behavioral, and physiological measures obtained from three assessments separated by substantial temporal intervals. Out results demonstrated that individual differences in emotion regulation ability, as indexed by an objective measure of emotional state, corrugator electromyography, predicted self-reported success while regulating pain. In both emotion and pain paradigms, the amygdala reflected regulatory success. Notably, we found that greater emotion regulation success was associated with greater change of amygdalar activity following pain regulation. Furthermore, individual differences in degree of amygdalar change following emotion regulation were a strong predictor of pain regulation success, as well as of the degree of amygdalar engagement following pain regulation. These findings suggest that common individual differences in emotion and pain regulatory success are reflected in a neural structure known to contribute to appraisal processes.

Simulated annealing technique for fast learning of SOM networks

The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for multidimensional input datasets. In this paper, we present an application of the simulated annealing procedure to the SOM learning algorithm with the aim to obtain a fast learning and better performances in terms of quantization error. The proposed learning algorithm is called Fast Learning Self-Organized Map, and it does not affect the easiness of the basic learning algorithm of the standard SOM. The proposed learning algorithm also improves the quality of resulting maps by providing better clustering quality and topology preservation of input multi-dimensional data. Several experiments are used to compare the proposed approach with the original algorithm and some of its modification and speed-up techniques.

New governance approaches to environmental regulation: an example of the Code for Sustainable Homes (CSH)

Environmental policy in the United Kingdom (UK) is witnessing a shift from command-and-control approaches towards more innovation-orientated environmental governance arrangements. These governance approaches are required which create institutions which support actors within a domain for learning not only about policy options, but also about their own interests and preferences. The need for construction actors to understand, engage and influence this process is critical to establishing policies which support innovation that satisfies each constituent’s needs. This capacity is particularly salient in an era where the expanding raft of environmental regulation is ushering in system-wide innovation in the construction sector. In this paper, the Code for Sustainable Homes (the Code) in the UK is used to demonstrate the emergence and operation of these new governance arrangements. The Code sets out a significant innovation challenge for the house-building sector with, for example, a requirement that all new houses must be zero-carbon by 2016. Drawing upon boundary organisation theory, the journey from the Code as a government aspiration, to the Code as a catalyst for the formation of the Zero Carbon Hub, a new institution, is traced and discussed. The case study reveals that the ZCH has demonstrated boundary organisation properties in its ability to be flexible to the needs and constraints of its constituent actors, yet robust enough to maintain and promote a common identity across regulation and industry boundaries.