FOWELL, S & LEVY, P 1995, 'DEVELOPING A NEW PROFESSIONAL PRACTICE - A MODEL FOR NETWORKED LEARNER SUPPORT IN HIGHER-EDUCATION', JOURNAL OF DOCUMENTATION, vol. 51, no. 3, pp. 271-280.
View/Download from: Publisher's site
FOWELL, SP & LEVY, P 1995, 'COMPUTER-MEDIATED COMMUNICATION IN THE INFORMATION CURRICULUM - AN INITIATIVE IN COMPUTER-SUPPORTED COLLABORATIVE LEARNING', EDUCATION FOR INFORMATION, vol. 13, no. 3, pp. 193-210.
Jay, CB 1995, 'A semantics for shape', Science of Computer Programming, vol. 25, pp. 251-283.
Jay, CB 1995, 'Covariant types'.
Jay, CB & Ghani, N 1995, 'The virtues of eta-expansion', J. of Functional Programming, vol. 5, pp. 135-154.
LIN, CJ & LIN, CT 1995, 'ADAPTIVE FUZZY CONTROL OF UNSTABLE NONLINEAR-SYSTEMS', INTERNATIONAL JOURNAL OF NEURAL SYSTEMS, vol. 6, no. 3, pp. 283-298.
View/Download from: Publisher's site
LIN, CT 1995, 'A NEURAL FUZZY CONTROL-SYSTEM WITH STRUCTURE AND PARAMETER LEARNING', FUZZY SETS AND SYSTEMS, vol. 70, no. 2-3, pp. 183-212.
View/Download from: Publisher's site
Lin, CT & Lee, CSG 1995, 'A Multi-Valued Boltzmann Machine', IEEE Transactions on Systems, Man, and Cybernetics, vol. 25, no. 4, pp. 660-669.
View/Download from: Publisher's site
View description>>
The idea of Hopfield network is based on the Ising spin glass model in which each spin has only two possible states: up and down. By introducing stochastic factors into this network and performing a simulated annealing process on it, it becomes a Boltzmann machine which can escape from local minimum states to achieve the global minimum. This paper generalizes the above ideas to multi-value case based on the XY spin glass model in which each spin can be in any direction in a plane. Simply using the gradient descent method and the analog Hopfield network, two different analog connectionist structures and their corresponding evolving rules are first designed to transform the XY spin glass model to distributed computational models. These two analog computational models are single-layered connectionist structures and multi-layered Hopfield analog networks. The latter network eases the node (neuron) computational requirement of the former at the expense of more neurons and connections. With the proposed evolving rules, the proposed models evolve according to a predefined Hamiltonian (energy function) which will decrease until it reaches a (perhaps local) minimum. Since these two structures can easily get stuck in local minima, a multivalued Boltzmann machine is proposed which adopts the discrete planar spin glass model for the local minimum problem. Each neuron in the multi-valued Boltzmann machine can only take n discrete directions (states). The stochastic simulated annealing method is introduced to the evolving rules of the multi-valued Boltzmann machine to solve the local minimum problem. The multi-valued Boltzmann machine can be applied to the mobile robot navigation problem by defining proper artificial magnetic field on the traverse terrain. This new artificial magnetic field approach for the mobile robot navigation problem has shown to have several advantages over existing graph search and potential field techniques. © 1995 IEEE
Lin, CT & Lu, YC 1995, 'A Neural Fuzzy System with Linguistic Teaching Signals', IEEE Transactions on Fuzzy Systems, vol. 3, no. 2, pp. 169-189.
View/Download from: Publisher's site
View description>>
A neural fuzzy system learning with linguistic teaching signals is proposed in this paper. This system is able to process and learn numerical information as well as linguistic information. It can be used either as an adaptive fuzzy expert system or as an adaptive fuzzy controller. At first, we propose a fivelayered neural network for the connectionist realization of a fuzzy inference system. The connectionist structure can house fuzzy logic rules and membership functions for fuzzy inference. We use a-level sets of fuzzy numbers to represent linguistic information. The inputs, outputs, and weights of the proposed network can be fuzzy numbers of any shape. Furthermore, they can be hybrid of fuzzy numbers and numerical numbers through the use of fuzzy singletons. Based on interval arithmetics, two kinds of learning schemes are developed for the proposed system: fuzzy supervised learning and fuzzy reinforcement learning. They extend the normal supervised and reinforcement learning techniques to the learning problems where only linguistic teaching signals are available. The fuzzy supervised learning scheme can train the proposed system with desired fuzzy input-output pairs which are fuzzy numbers instead of the normal numerical values. With fuzzy supervised learning, the proposed system can be used for rule base concentration to reduce the number of rules in a fuzzy rule base. In the fuzzy reinforcement learning problem that we consider, the reinforcement signal from the environment is linguistic information (fuzzy critic signal) such as “good,” “very good,” or “bad,” instead of the normal numerical critic values such as “0" (success) or 1" (failure). With the fuzzy critic signal from the environment, the proposed system can learn proper fuzzy control rules and membership functions. We discuss two kinds of fuzzy reinforcement learning problems: singlestep prediction problems and multistep prediction problems. Simulation results are presented to illustrate the performa...
Lin, CT, Lin, CJ & Lee, CSG 1995, 'Fuzzy adaptive learning control network with on-line neural learning', Fuzzy Sets and Systems, vol. 71, no. 1, pp. 25-45.
View/Download from: Publisher's site
View description>>
This paper addresses the structure and the associated on-line learning algorithms of a feedforward multilayered connectionist network for realizing the basic elements and functions of a traditional fuzzy logic controller. The proposed Fuzzy Adaptive Learning COntrol Network (FALCON) can be contrasted with the traditional fuzzy logic control systems in their network structure and learning ability. The connectionist structure of the proposed FALCON can be constructed from training examples by neural learning techniques to find proper fuzzy partitions, membership functions, and fuzzy logic rules. Two complementary on-line structure/parameter learning algorithms, FALCON-FSM and FALCON-ART, are proposed for constructing the FALCON dynamically. The FALCON-FSM combines the backpropagation learning scheme for parameter learning and a fuzzy similarity measure for structure learning. The FALCON-FSN can find proper fuzzy logic rules, membership functions, and the size of output partitions simultaneously. In the FALCON-FSM algorithm, the input and output spaces are partitioned into "grids". The grid-typed space partitioning certainly makes both the fuzzy logic controller software emulation and fuzzy chip implementation convenient. However, as the number of input/output variables increases, the number of partitioned grids will grow combinatorially. To avoid the problem of combinatorial growth of partitioned grids in some complex systems, the FALCON-ART algorithm is developed, which can partition the input and output spaces in a more flexible way based on the distribution of the training data. The FALCON-ART combines the backpropagation learning scheme for parameter learning and a fuzzy ART algorithm for structure learning. The FALCON-ART can on-line partition the input and output spaces, tune membership functions and find proper fuzzy logic rules dynamically. Computer simulations were conducted to illustrate the performance and applicability of both FALCON-FSM and FALCON-ART ...
Peppas, P & Williams, MA 1995, 'Constructive modelings for theory change', Notre Dame Journal of Formal Logic, vol. 36, no. 1, pp. 120-133.
View/Download from: Publisher's site
View description>>
Alchourron, Gärdenfors and Makinson have developed and investigated a set of rationality postulates which appear to capture much of what is required of any rational system of theory revision. This set of postulates describes a class of revision functions, however it does not provide a constructive way of defining such a function. There are two principal constructions of revision functions, namely an epistemic entrenchment and a system of spheres. We refer to their approach as the AGM paradigm. We provide a new constructive modeling for a revision function based on a nice preorder on models, and furthermore we give explicit conditions under which a nice preorder on models, an epistemic entrenchment, and a system of spheres yield the same revision function. Moreover, we provide an identity which captures the relationship between revision functions and update operators (as defined by Katsuno and Mendelzon). © 1995, Duke University Press. All Rights Reserved.
PFEIFFER, M & LEANEY, J 1995, 'THE SIMPLE RELIABLE MONITOR - A FORMALIZATION OF THE CONCEPT OF A SAFE SOFTWARE MONITOR', AUSTRALIAN COMPUTER JOURNAL, vol. 27, no. 1, pp. 9-15.
Wang, J, Xu, G & Wang, N 1995, 'Mathematical model for calculating the flux of laser scattering by single particle in arbitrary directions', Yingyong Jiguang/Applied Laser Technology, vol. 15, no. 2, pp. 79-78.
View description>>
A formula for calculating the scattered light intensity from a single particle by using the Mie's theory was derived here, and based on which, the scattered light-flux in a certain solid-angle in arbitrary direction was further obtained to correct the mistake occuring in the expression on the light-flux existed in one of the references listed.
Chang, HT, Shyu, JM, Lin, CT, Chen, OTC, Deng, HJ, Chen, WJ, Luo, SR, Hsu, YR, Lu, YC & Shyu, HC 1995, 'Pipelined fuzzy reasoning processor with software development system and its application on the crane control problem', Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, pp. 3658-3665.
View description>>
Fuzzy reasoning processors have been employed in many commercial and industrial applications. A high-performance pipelined, single-instruction-stream, and single-data-stream architecture of fuzzy reasoning engine has been designed. Based on this architecture, the proposed VLSI processor for embedded real-time fuzzy logic applications was fabricated in a 0.8-mm CMOS technology. Its computation power can reach 2.5 Million Fuzzy Logic Inferences Per Second (MFLIPS) at a system clock of 20 MHz. In order to efficiently realize fuzzy applications, a software development system under Microsoft Windows has also been designed. For the purpose of showing its effectiveness, we have carried out the experiments to successfully control the crane system. Experimental results have shown the efficient cooperative control of the proposed fuzzy reasoning processor with software development system and can be contrast with conventional computer software interface control.
Jay, CB 1995, 'Polynomial Polymorph\-ism', Proceedings of the Eighteenth Australasian Computer Science Conference: Glenelg, South Australia 1–3 February, 1995, ACS Communications, pp. 237-243.
Jay, CB 1995, 'Shape Analysis for Parallel Computing', Proceedings of the fourth international parallel computing workshop: Imperial College London, 25–26 September, 1995, Imperial College/Fujitsu Parallel Computing Research Centre, pp. 287-298.
LIN, CJ & LIN, CT 1995, 'REINFORCEMENT LEARNING FOR ART-BASED FUZZY ADAPTIVE LEARNING CONTROL NETWORKS', PROCEEDINGS OF 1995 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS, VOLS I-IV, 4th IEEE International Conference on Fuzzy Systems/2nd International Fuzzy Engineering Symposium (FUZZY-IEEE/IFES 95), I E E E, YOKOHAMA, JAPAN, pp. 1299-1306.
LIN, CT, LIN, CJ & CHUNG, IF 1995, 'Neural fuzzy control of unstable nonlinear systems', 1995 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS, VOLS 1-5, 1995 IEEE International Conference on Systems, Man and Cybernetics - Intelligent Systems for the 21st-Century, I E E E, VANCOUVER, CANADA, pp. 3666-3671.
Lister, R & Stone, JV 1995, 'Empirical study of the time complexity of various error functions with conjugate gradient back propagation', IEEE International Conference on Neural Networks - Conference Proceedings, pp. 237-241.
View description>>
We describe an empirical comparison of the scaling behaviour of six error functions, on a conjugate gradient form of Back Propagation. We classify the functions according to the limit behaviours of their respective error signals, as the target value and the actual output value approach opposite extremes. These limit behaviours are zero limit, finite limit, and infinite limit. Despite such a wide divergence in their limit behaviours, we find that all six error functions exhibit a median run-time order of approximately O(N 4) on the N-2-N encoder. This result indicates that, while some factors affecting the scaling behaviour of standard and conjugate gradient Back Propagation have been previously identified (such as saturation), other factors remain unidentified.
Lister, R & Stone, JV 1995, 'Error functions, error signals, and conjugate gradient back propagation', IEE Conference Publication, pp. 76-81.
View description>>
We propose a three-class taxonomy of error functions, based on the limit behaviour of the error signal. We classify four established error functions: the quadratic, Fahlman's Quickprop, entropy, and the exception error function. We introduce two new error functions, and benchmark all six on the N-2-N encoder. The two new functions found correct solutions faster and more reliably than the established functions.