No abstract

A formal selection and pruning technique based on the concept of local relative sensitivity index is proposed for feedforward artificial neural networks. The mechanism of backpropagation training algorithm is revisited and the theoretical foundation of the improved selection and pruning technique is presented. This technique is based on parallel pruning of weights which are relatively redundant in a subgroup of a feedforward neural network. Comparative studies with a similar technique proposed in the literature show that the improved technique provides better pruning results in terms of reduction of model residues, improvement of generalization capability and reduction of network complexity. The effectiveness of the improved technique is demonstrated in developing neural network (NN) models of a number of nonlinear systems including three bit parity problem, Van der Pol equation, a chemical processes and two nonlinear discrete-time systems using the backpropagation training algorithm with adaptive learning rate.

Study of Engineering Control is often perceived by students as highly complex and a difficult hurdle to pass. The authors of this paper were prompted to propose changes to the teaching of a final year control unit of a BEng (Hons) Degree in Electrical and Electronic Engineering after numerous students had expressed concern about their ability to be successful in this subject. The average exam grade had also been low for several years. Teaching previously had been didactic in nature, based on lectures, predefined laboratory exercises and tutorial sessions. Use of appropriate based workshop study sessions can lead to higher levels of attainment and development, and student perception is that the experience is informal, engaging but appropriately challenging. A series of workshop sessions were introduced into term two of the unit teaching over a period of several weeks. Students worked individually on learning improvement activities, within cooperative groups in an informal workshop environment. The tasks set to the students were challenging and designed to improve general skills in Control, and students were encouraged to explore further areas that interested them. Results presented in this paper describe an improvement in student engagement, confidence in exam preparation and communication of learning measured by grades since the introduction of the workshop sessions, and improved feedback from students. The workshop sessions were integrated into the teaching scheme rather than being elective, and did not require any additional timetable sessions.

No abstract

No abstract