Wang Dan. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2001. / Includes bibliographical references (p. 128-138). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web. / Abstracts in English and Chinese.

by Chen, Zhong-Yu. / Thesis (Ph.D.)--Chinese University of Hong Kong, 1996. / Includes bibliographical references (p. 146-[154]). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Mode of access: World Wide Web.

by Lo Wai Kit. / Thesis (M.Phil.)--Chinese University of Hong Kong, 1996. / Includes bibliographical references (leaves 151-160). / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Applications of Speech Synthesis --- p.2 / Chapter 1.1.1 --- Human Machine Interface --- p.2 / Chapter 1.1.2 --- Speech Aids --- p.3 / Chapter 1.1.3 --- Text-To-Speech (TTS) system --- p.4 / Chapter 1.1.4 --- Speech Dialogue System --- p.4 / Chapter 1.2 --- Current Status in Speech Synthesis --- p.6 / Chapter 1.2.1 --- Concatenation Based --- p.6 / Chapter 1.2.2 --- Parametric Based --- p.7 / Chapter 1.2.3 --- Articulatory Based --- p.7 / Chapter 1.2.4 --- Application of Neural Network in Speech Synthesis --- p.8 / Chapter 1.3 --- The Proposed Neural Network Speech Synthesis --- p.9 / Chapter 1.3.1 --- Motivation --- p.9 / Chapter 1.3.2 --- Objectives --- p.9 / Chapter 1.4 --- Thesis outline --- p.11 / Chapter 2 --- Linguistic Basics for Speech Synthesis --- p.12 / Chapter 2.1 --- Relations between Linguistic and Speech Synthesis --- p.12 / Chapter 2.2 --- Basic Phonology and Phonetics --- p.14 / Chapter 2.2.1 --- Phonology --- p.14 / Chapter 2.2.2 --- Phonetics --- p.15 / Chapter 2.2.3 --- Prosody --- p.16 / Chapter 2.3 --- Transcription Systems --- p.17 / Chapter 2.3.1 --- The Employed Transcription System --- p.18 / Chapter 2.4 --- Cantonese Phonology --- p.20 / Chapter 2.4.1 --- Some Properties of Cantonese --- p.20 / Chapter 2.4.2 --- Initial --- p.21 / Chapter 2.4.3 --- Final --- p.23 / Chapter 2.4.4 --- Lexical Tone --- p.25 / Chapter 2.4.5 --- Variations --- p.26 / Chapter 2.5 --- The Vowel Quadrilaterals --- p.29 / Chapter 3 --- Speech Synthesis Technology --- p.32 / Chapter 3.1 --- The Human Speech Production --- p.32 / Chapter 3.2 --- Important Issues in Speech Synthesis System --- p.34 / Chapter 3.2.1 --- Controllability --- p.34 / Chapter 3.2.2 --- Naturalness --- p.34 / Chapter 3.2.3 --- Complexity --- p.35 / Chapter 3.2.4 --- Information Storage --- p.35 / Chapter 3.3 --- Units for Synthesis --- p.37 / Chapter 3.4 --- Type of Synthesizer --- p.40 / Chapter 3.4.1 --- Copy Concatenation --- p.40 / Chapter 3.4.2 --- Vocoder --- p.41 / Chapter 3.4.3 --- Articulatory Synthesis --- p.44 / Chapter 4 --- Neural Network Speech Synthesis with Articulatory Control --- p.47 / Chapter 4.1 --- Neural Network Approximation --- p.48 / Chapter 4.1.1 --- The Approximation Problem --- p.48 / Chapter 4.1.2 --- Network Approach for Approximation --- p.49 / Chapter 4.2 --- Artificial Neural Network for Phone-based Speech Synthesis --- p.53 / Chapter 4.2.1 --- Network Approximation for Speech Signal Synthesis --- p.53 / Chapter 4.2.2 --- Feed forward Backpropagation Neural Network --- p.56 / Chapter 4.2.3 --- Radial Basis Function Network --- p.58 / Chapter 4.2.4 --- Parallel Operating Synthesizer Networks --- p.59 / Chapter 4.3 --- Template Storage and Control for the Synthesizer Network --- p.61 / Chapter 4.3.1 --- Implicit Template Storage --- p.61 / Chapter 4.3.2 --- Articulatory Control Parameters --- p.61 / Chapter 4.4 --- Summary --- p.65 / Chapter 5 --- Prototype Implementation of the Synthesizer Network --- p.66 / Chapter 5.1 --- Implementation of the Synthesizer Network --- p.66 / Chapter 5.1.1 --- Network Architectures --- p.68 / Chapter 5.1.2 --- Spectral Templates for Training --- p.74 / Chapter 5.1.3 --- System requirement --- p.76 / Chapter 5.2 --- Subjective Listening Test --- p.79 / Chapter 5.2.1 --- Sample Selection --- p.79 / Chapter 5.2.2 --- Test Procedure --- p.81 / Chapter 5.2.3 --- Result --- p.83 / Chapter 5.2.4 --- Analysis --- p.86 / Chapter 5.3 --- Summary --- p.88 / Chapter 6 --- Simplified Articulatory Control for the Synthesizer Network --- p.89 / Chapter 6.1 --- Coarticulatory Effect in Speech Production --- p.90 / Chapter 6.1.1 --- Acoustic Effect --- p.90 / Chapter 6.1.2 --- Prosodic Effect --- p.91 / Chapter 6.2 --- Control in various Synthesis Techniques --- p.92 / Chapter 6.2.1 --- Copy Concatenation --- p.92 / Chapter 6.2.2 --- Formant Synthesis --- p.93 / Chapter 6.2.3 --- Articulatory synthesis --- p.93 / Chapter 6.3 --- Articulatory Control Model based on Vowel Quad --- p.94 / Chapter 6.3.1 --- Modeling of Variations with the Articulatory Control Model --- p.95 / Chapter 6.4 --- Voice Correspondence : --- p.97 / Chapter 6.4.1 --- For Nasal Sounds ´ؤ Inter-Network Correspondence --- p.98 / Chapter 6.4.2 --- In Flat-Tongue Space - Intra-Network Correspondence --- p.101 / Chapter 6.5 --- Summary --- p.108 / Chapter 7 --- Pause Duration Properties in Cantonese Phrases --- p.109 / Chapter 7.1 --- The Prosodic Feature - Inter-Syllable Pause --- p.110 / Chapter 7.2 --- Experiment for Measuring Inter-Syllable Pause of Cantonese Phrases --- p.111 / Chapter 7.2.1 --- Speech Material Selection --- p.111 / Chapter 7.2.2 --- Experimental Procedure --- p.112 / Chapter 7.2.3 --- Result --- p.114 / Chapter 7.3 --- Characteristics of Inter-Syllable Pause in Cantonese Phrases --- p.117 / Chapter 7.3.1 --- Pause Duration Characteristics for Initials after Pause --- p.117 / Chapter 7.3.2 --- Pause Duration Characteristic for Finals before Pause --- p.119 / Chapter 7.3.3 --- General Observations --- p.119 / Chapter 7.3.4 --- Other Observations --- p.121 / Chapter 7.4 --- Application of Pause-duration Statistics to the Synthesis System --- p.124 / Chapter 7.5 --- Summary --- p.126 / Chapter 8 --- Conclusion and Further Work --- p.127 / Chapter 8.1 --- Conclusion --- p.127 / Chapter 8.2 --- Further Extension Work --- p.130 / Chapter 8.2.1 --- Regularization Network Optimized on ISD --- p.130 / Chapter 8.2.2 --- Incorporation of Non-Articulatory Parameters to Control Space --- p.130 / Chapter 8.2.3 --- Experiment on Other Prosodic Features --- p.131 / Chapter 8.2.4 --- Application of Voice Correspondence to Cantonese Coda Discrim- ination --- p.131 / Chapter A --- Cantonese Initials and Finals --- p.132 / Chapter A.1 --- Tables of All Cantonese Initials and Finals --- p.132 / Chapter B --- Using Distortion Measure as Error Function in Neural Network --- p.135 / Chapter B.1 --- Formulation of Itakura-Saito Distortion Measure for Neural Network Error Function --- p.135 / Chapter B.2 --- Formulation of a Modified Itakura-Saito Distortion (MISD) Measure for Neural Network Error Function --- p.137 / Chapter C --- Orthogonal Least Square Algorithm for RBFNet Training --- p.138 / Chapter C.l --- Orthogonal Least Squares Learning Algorithm for Radial Basis Function Network Training --- p.138 / Chapter D --- Phrase Lists --- p.140 / Chapter D.1 --- Two-Syllable Phrase List for the Pause Duration Experiment --- p.140 / Chapter D.1.1 --- 兩字詞 --- p.140 / Chapter D.2 --- Three/Four-Syllable Phrase List for the Pause Duration Experiment --- p.144 / Chapter D.2.1 --- 片語 --- p.144

Ng Chi-fai. / Thesis submitted in: December 2000. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2001. / Includes bibliographical references (leaves 69-73). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgments --- p.iii / List of Figures --- p.viii / List of Tables --- p.ix / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation and Objectives --- p.1 / Chapter 1.2 --- Principles of Feedforward Neural Network Approximation --- p.1 / Chapter 1.3 --- Contribution of The Thesis --- p.5 / Chapter 1.4 --- Outline of The Thesis --- p.5 / Chapter 2 --- Feedforward Neural Networks: An Approximator for Nonlinear Control Law --- p.8 / Chapter 2.1 --- Optimization Methods Applied in Feedforward Neural Network Approximation --- p.8 / Chapter 2.2 --- Example in Supervised Learning --- p.10 / Chapter 2.2.1 --- Problem Description --- p.10 / Chapter 2.2.2 --- Neural Network Configuration and Training --- p.12 / Chapter 2.2.3 --- Simulation Result --- p.13 / Chapter 3 --- Neural Based Approximation of Center Manifold Equations --- p.19 / Chapter 3.1 --- Solving Center Manifold Equations by Feedforward Neural Network Approx- imation --- p.19 / Chapter 3.2 --- Example --- p.21 / Chapter 3.2.1 --- Problem Description --- p.21 / Chapter 3.2.2 --- Simulation Result --- p.24 / Chapter 3.2.3 --- Discussion --- p.24 / Chapter 4 --- Connection of Center Manifold Equations to Output Regulation Problem --- p.29 / Chapter 4.1 --- Output Regulation Theory --- p.29 / Chapter 4.2 --- Reduction of Regulator Equation into Center Manifold Equations --- p.31 / Chapter 5 --- Application to the Control Design of Ball and Beam System --- p.34 / Chapter 5.1 --- Problem Description --- p.34 / Chapter 5.2 --- Neural Approximation Solution of Center Manifold Equations --- p.37 / Chapter 5.3 --- Simulation Results --- p.38 / Chapter 5.4 --- Discussion --- p.45 / Chapter 6 --- Neural Based Disturbance Rejection of Nonlinear Benchmark Problem (TORA System) --- p.48 / Chapter 6.1 --- Problem Description --- p.48 / Chapter 6.2 --- Neural based Approximation of the Center Manifold Equations of TORA System --- p.51 / Chapter 6.3 --- Simulation Results --- p.53 / Chapter 6.4 --- Discussion --- p.59 / Chapter 7 --- Conclusion --- p.62 / Chapter 7.1 --- Future Works --- p.63 / Chapter A --- Center Manifold Theory --- p.64 / Chapter B --- Relation between Center Manifold Equation and Output Regulation Prob- lem --- p.66 / Biography --- p.68 / References --- p.69

I conduct a study of learning in HMAX-like models, which are hierarchical models of visual processing in biological vision systems. Such models compute a new representation for an image based on the similarity of image sub-parts to a number of specific patterns, called prototypes. Despite being a central piece of the overall model, the issue of choosing the best prototypes for a given task is still an open problem. I study this problem, and consider the best way to increase task performance while decreasing the computational costs of the model. This work broadens our understanding of HMAX and related hierarchical models as tools for theoretical neuroscience, while simultaneously increasing the utility of such models as applied computer vision systems.

Object recognition in video has seen giant strides in accuracy improvements in the last few years, a testament to the computational capacity of deep convolutional neural networks. However, this computational capacity of software-based neural networks coincides with high power consumption compared to that of some spiking neural networks (SNNs), up to 300,000 times more energy per synaptic event in IBM's TrueNorth chip, for example. SNNs are also well-suited to exploit the precise timing of event-driven image sensors, which transmit asynchronous "events" only when the luminance of a pixel changes above or below a threshold value. The combination of event-based imagers and SNNs becomes a straightforward way to achieve low power consumption in object recognition tasks. This thesis compares different linear classifiers for two low-power, hardware-friendly, spiking, unsupervised neural network architectures, SSLCA and HFirst, in response to asynchronous event-based data, and explores their ability to learn and recognize patterns from two event-based image datasets, N-MNIST and CIFAR10-DVS. By performing a grid search of important SNN and classifier hyperparameters, we also explore how to improve classification performance of these architectures. Results show that a softmax regression classifier exhibits modest accuracy gains (0.73%) over the next-best performing linear support vector machine (SVM), and considerably outperforms a single layer perceptron (by 5.28%) when classification performance is averaged over all datasets and spiking neural network architectures with varied hyperparameters. Min-max normalization of the inputs to the linear classifiers aides in classification accuracy, except in the case of the single layer perceptron classifier. We also see the highest reported classification accuracy for spiking convolutional networks on N-MNIST and CIFAR10-DVS, increasing this accuracy from 97.77% to 97.82%, and 29.67% to 31.76%, respectively. These findings are relevant for any system employing unsupervised SNNs to extract redundant features from event-driven data for recognition.

Human balance control is a complex feedback system that must be adaptable and robust in an infinitely varying external environment. It is probable that there are many concurrent control loops occurring in the central nervous system that achieve stability for a variety of postural perturbations. Though many engineering models of human balance control have been tested, no models of how these controllers might operate within the nervous system have yet been developed. We have focused on building a model of a proprioceptive feedback loop with simulated neurons. The proprioceptive referenced portion of human balance control has been successfully modeled by a PD controller with a time delay and output torque positive feedback. For this model, angular position is measured at the ankle and corrective torque is applied about the joint to maintain a vertical orientation. In this paper, we construct a neural network that performs addition, subtraction, multiplication, differentiation and signal filtering to demonstrate that a simulated biological neural system based off of the engineering control model is capable of matching human test subject dynamics.

An algorithm for solving the large-scale Traveling Salesman Problem is presented. Research into past work in the area of Hopfield neural network use in solving the Traveling Salesman Problem has yielded design ideas that have been incorporated into this work. The algorithm consists of an unsupervised learning algorithm and a recursive Hopfield neural network. The unsupervised learning algorithm was used to decompose the problem into clusters. The recursive Hopfield neural network was applied to the centroids of the clusters, then to the cities in each cluster, in order to find an optimal path. An improvement in both computation speed and solution accuracy is shown by the proposed algorithm over the straight use of the Hopfield neural network.

Abstract not available

A new Bayesian framework for training and selecting the complexity of artificial neural networks (ANNs) is developed in this thesis, based on Markov chain Monte Carlo (MCMC) techniques. The primary motivation of the research presented is the incorporation of uncertainty into ANNs used for water resources modelling, with emphasis placed on obtaining accurate results, while maintaining simplicity of implementation, which is considered to be of utmost importance for adoption of the framework by practitioners in this field. By applying the Bayesian framework to a number of synthetic and real-world case studies and by comparison with a state-of-the-art ANN development approach, it is shown throughout this thesis how the Bayesian approach can be used to address the three most significant issues facing the wider acceptance of ANNs in this field; namely generalisability, interpretability and uncertainty. The state-of-the-art approach is devised through reviewing and, where necessary, improving current best practice deterministic ANN development methods, leading to the recommended use of the global SCE-UA optimisation algorithm, which has not been used before for ANN training, and the development of a modified connection weight approach for extracting knowledge from trained ANNs. The real-world case studies used in this research, which involve salinity forecasting in the River Murray at Murray Bridge, South Australia, and the forecasting of cyanobacteria (Anabaena spp.) in the River Murray at Morgan, South Australia, are used to demonstrate the practical value of the Bayesian framework, particularly when extrapolation is required and when the available data are of poor quality. These issues lead to poor model performance when deterministic ANN development methods are applied, yet as the generated predictions are deterministic, there is no direct way of assessing their quality. Application of the proposed Bayesian framework leads to better average performance of the ANN models developed, since a minimal ANN structure is selected and a more generalised input-output mapping is obtained. More importantly, prediction limits are provided which quantify the uncertainty in the predictions and enable management and design decisions to be made based on a known level of confidence. / http://proxy.library.adelaide.edu.au/login?url= http://library.adelaide.edu.au/cgi-bin/Pwebrecon.cgi?BBID=1235735 / Thesis (Ph.D.) --, University of Adelaide, School of Civil and Environmental Engineering, 2006