Mixture density network training by computation in parameter space

Training Mixture Density Network (MDN) configurations within the NETLAB framework takes time due to the nature of the computation of the error function and the gradient of the error function. By optimising the computation of these functions, so that gradient information is computed in parameter space, training time is decreased by at least a factor of sixty for the example given. Decreased training time increases the spectrum of problems to which MDNs can be practically applied making the MDN framework an attractive method to the applied problem solver.

Inventory parameter management and focused continuous improvement for repetitive batch manufacturers

What this thesis proposes is a methodology to assist repetitive batch manufacturers in the adoption of certain aspects of the Lean Production principles. The methodology concentrates on the reduction of inventory through the setting of appropriate batch sizes, taking account of the effect of sequence dependent set-ups and the identification and elimination of bottlenecks. It uses a simple Pareto and modified EBQ based analysis technique to allocate items to period order day classes based on a combination of each item's annual usage value and set-up cost. The period order day classes the items are allocated to are determined by the constraints limits in the three measured dimensions, capacity, administration and finance. The methodology overcomes the limitations associated with MRP in the area of sequence dependent set-ups, and provides a simple way of setting planning parameters taking this effect into account by concentrating on the reduction of inventory through the systematic identification and elimination of bottlenecks through set-up reduction processes, so allowing batch sizes to reduce. It aims to help traditional repetitive batch manufacturers in a route to continual improvement by: Highlighting those areas where change would bring the greatest benefits. Modelling the effect of proposed changes. Quantifying the benefits that could be gained through implementing the proposed changes. Simplifying the effort required to perform the modelling process. It concentrates on increasing flexibility through managed inventory reduction through rationally decreasing batch sizes, taking account of sequence dependent set-ups and the identification and elimination of bottlenecks. This was achieved through the development of a software modelling tool, and validated through a case study approach.

Q parameter estimation using numerical simulations for linear and nonlinear transmission systems

We compare the Q parameter obtained from scalar, semi-analytical and full vector models for realistic transmission systems. One set of systems is operated in the linear regime, while another is using solitons at high peak power. We report in detail on the different results obtained for the same system using different models. Polarisation mode dispersion is also taken into account and a novel method to average Q parameters over several independent simulation runs is described. © 2006 Elsevier B.V. All rights reserved.

Semi-analytical Q parameter estimate in linear and nonlinear transmission systems

We compare the Q parameter obtained from the semi-analytical model with scalar and vector models for two realistic transmission systems. First a linear system with a compensated dispersion map and second a soliton transmission system.

Numerical Q parameter estimates for scalar and vector models in optical communication system simulations

The Q parameter scales differently with the noise power for the signal-noise and the noise-noise beating terms in scalar and vector models. Some procedures for including noise in the scalar model largely under-estimate the Q parameter. We propose a simple method for including noise within a scalar model which will allow both the noise-noise dominated limit and the signal-noise dominated limit to be treated consistently. © 2005 Elsevier B.V. All rights reserved.

Q parameter scaling in scalar and vector models with dominant signal-noise and noise-noise beating terms

The Q parameter scales differently with the noise power for the signal-noise and the noise-noise beating terms in scalar and vector models. Some procedures for including noise in the scalar model largely under-estimate the Q parameter.

The accuracy of parameter estimation from noisy data, with application to resonance peak estimation in distributed Brillouin sensing

Distributed Brillouin sensing of strain and temperature works by making spatially resolved measurements of the position of the measurand-dependent extremum of the resonance curve associated with the scattering process in the weakly nonlinear regime. Typically, measurements of backscattered Stokes intensity (the dependent variable) are made at a number of predetermined fixed frequencies covering the design measurand range of the apparatus and combined to yield an estimate of the position of the extremum. The measurand can then be found because its relationship to the position of the extremum is assumed known. We present analytical expressions relating the relative error in the extremum position to experimental errors in the dependent variable. This is done for two cases: (i) a simple non-parametric estimate of the mean based on moments and (ii) the case in which a least squares technique is used to fit a Lorentzian to the data. The question of statistical bias in the estimates is discussed and in the second case we go further and present for the first time a general method by which the probability density function (PDF) of errors in the fitted parameters can be obtained in closed form in terms of the PDFs of the errors in the noisy data.

Realisation of an effective dual-parameter sensor employing a single fibre Bragg grating structure

A novel technology for simultaneous and independent measurement of dual parameters is proposed and experimented. The length of a single fibre Bragg grating (FBG) is divided into two parts. The temperature variation and another measurand can be measured independently and simultaneously, and the thermal effect can be erased with great ease.

An investigation into inductive parameter learning in complex hierarchical knowledge structures representing clinical expertise

This dissertation investigates the very important and current problem of modelling human expertise. This is an apparent issue in any computer system emulating human decision making. It is prominent in Clinical Decision Support Systems (CDSS) due to the complexity of the induction process and the vast number of parameters in most cases. Other issues such as human error and missing or incomplete data present further challenges. In this thesis, the Galatean Risk Screening Tool (GRiST) is used as an example of modelling clinical expertise and parameter elicitation. The tool is a mental health clinical record management system with a top layer of decision support capabilities. It is currently being deployed by several NHS mental health trusts across the UK. The aim of the research is to investigate the problem of parameter elicitation by inducing them from real clinical data rather than from the human experts who provided the decision model. The induced parameters provide an insight into both the data relationships and how experts make decisions themselves. The outcomes help further understand human decision making and, in particular, help GRiST provide more accurate emulations of risk judgements. Although the algorithms and methods presented in this dissertation are applied to GRiST, they can be adopted for other human knowledge engineering domains.