The 'moving targets' training algorithm

A simple method for training the dynamical behavior of a neural network is derived. It is applicable to any training problem in discrete-time networks with arbitrary feedback. The method resembles back-propagation in that it is a least-squares, gradient-based optimization method, but the optimization is carried out in the hidden part of state space instead of weight space. A straightforward adaptation of this method to feedforward networks offers an alternative to training by conventional back-propagation. Computational results are presented for simple dynamical training problems, with varied success. The failures appear to arise when the method converges to a chaotic attractor. A patch-up for this problem is proposed. The patch-up involves a technique for implementing inequality constraints which may be of interest in its own right.

Capacity of the upstart algorithm

The storage capacity of multilayer networks with overlapping receptive fields is investigated for a constructive algorithm within a one-step replica symmetry breaking (RSB) treatment. We find that the storage capacity increases logarithmically with the number of hidden units K without saturating the Mitchison-Durbin bound. The slope of the logarithmic increase decays exponentionally with the stability with which the patterns have been stored.

Topographic mappings and feed-forward neural networks

This thesis is a study of the generation of topographic mappings - dimension reducing transformations of data that preserve some element of geometric structure - with feed-forward neural networks. As an alternative to established methods, a transformational variant of Sammon's method is proposed, where the projection is effected by a radial basis function neural network. This approach is related to the statistical field of multidimensional scaling, and from that the concept of a 'subjective metric' is defined, which permits the exploitation of additional prior knowledge concerning the data in the mapping process. This then enables the generation of more appropriate feature spaces for the purposes of enhanced visualisation or subsequent classification. A comparison with established methods for feature extraction is given for data taken from the 1992 Research Assessment Exercise for higher educational institutions in the United Kingdom. This is a difficult high-dimensional dataset, and illustrates well the benefit of the new topographic technique. A generalisation of the proposed model is considered for implementation of the classical multidimensional scaling (¸mds}) routine. This is related to Oja's principal subspace neural network, whose learning rule is shown to descend the error surface of the proposed ¸mds model. Some of the technical issues concerning the design and training of topographic neural networks are investigated. It is shown that neural network models can be less sensitive to entrapment in the sub-optimal global minima that badly affect the standard Sammon algorithm, and tend to exhibit good generalisation as a result of implicit weight decay in the training process. It is further argued that for ideal structure retention, the network transformation should be perfectly smooth for all inter-data directions in input space. Finally, there is a critique of optimisation techniques for topographic mappings, and a new training algorithm is proposed. A convergence proof is given, and the method is shown to produce lower-error mappings more rapidly than previous algorithms.

Magnification factors for the GTM algorithm

The Generative Topographic Mapping (GTM) algorithm of Bishop et al. (1997) has been introduced as a principled alternative to the Self-Organizing Map (SOM). As well as avoiding a number of deficiencies in the SOM, the GTM algorithm has the key property that the smoothness properties of the model are decoupled from the reference vectors, and are described by a continuous mapping from a lower-dimensional latent space into the data space. Magnification factors, which are approximated by the difference between code-book vectors in SOMs, can therefore be evaluated for the GTM model as continuous functions of the latent variables using the techniques of differential geometry. They play an important role in data visualization by highlighting the boundaries between data clusters, and are illustrated here for both a toy data set, and a problem involving the identification of crab species from morphological data.

A scatterometer neural network sensor model with input noise

The ERS-1 satellite carries a scatterometer which measures the amount of radiation scattered back toward the satellite by the ocean's surface. These measurements can be used to infer wind vectors. The implementation of a neural network based forward model which maps wind vectors to radar backscatter is addressed. Input noise cannot be neglected. To account for this noise, a Bayesian framework is adopted. However, Markov Chain Monte Carlo sampling is too computationally expensive. Instead, gradient information is used with a non-linear optimisation algorithm to find the maximum em a posteriori probability values of the unknown variables. The resulting models are shown to compare well with the current operational model when visualised in the target space.

A scatterometer neural network sensor model with input noise

The ERS-1 satellite carries a scatterometer which measures the amount of radiation scattered back toward the satellite by the ocean's surface. These measurements can be used to infer wind vectors. The implementation of a neural network based forward model which maps wind vectors to radar backscatter is addressed. Input noise cannot be neglected. To account for this noise, a Bayesian framework is adopted. However, Markov Chain Monte Carlo sampling is too computationally expensive. Instead, gradient information is used with a non-linear optimisation algorithm to find the maximum em a posteriori probability values of the unknown variables. The resulting models are shown to compare well with the current operational model when visualised in the target space.

A combined neural network and DEA for measuring efficiency of large scale datasets

Data Envelopment Analysis (DEA) is one of the most widely used methods in the measurement of the efficiency and productivity of Decision Making Units (DMUs). DEA for a large dataset with many inputs/outputs would require huge computer resources in terms of memory and CPU time. This paper proposes a neural network back-propagation Data Envelopment Analysis to address this problem for the very large scale datasets now emerging in practice. Neural network requirements for computer memory and CPU time are far less than that needed by conventional DEA methods and can therefore be a useful tool in measuring the efficiency of large datasets. Finally, the back-propagation DEA algorithm is applied to five large datasets and compared with the results obtained by conventional DEA.

A novel silicon membrane-based biosensing platform using distributive sensing strategy and artificial neural networks for feature analysis

A novel biosensing system based on a micromachined rectangular silicon membrane is proposed and investigated in this paper. A distributive sensing scheme is designed to monitor the dynamics of the sensing structure. An artificial neural network is used to process the measured data and to identify cell presence and density. Without specifying any particular bio-application, the investigation is mainly concentrated on the performance testing of this kind of biosensor as a general biosensing platform. The biosensing experiments on the microfabricated membranes involve seeding different cell densities onto the sensing surface of membrane, and measuring the corresponding dynamics information of each tested silicon membrane in the form of a series of frequency response functions (FRFs). All of those experiments are carried out in cell culture medium to simulate a practical working environment. The EA.hy 926 endothelial cell lines are chosen in this paper for the bio-experiments. The EA.hy 926 endothelial cell lines represent a particular class of biological particles that have irregular shapes, non-uniform density and uncertain growth behaviour, which are difficult to monitor using the traditional biosensors. The final predicted results reveal that the methodology of a neural-network based algorithm to perform the feature identification of cells from distributive sensory measurement has great potential in biosensing applications.

Evolutionary neural networks : models and applications

The scaling problems which afflict attempts to optimise neural networks (NNs) with genetic algorithms (GAs) are disclosed. A novel GA-NN hybrid is introduced, based on the bumptree, a little-used connectionist model. As well as being computationally efficient, the bumptree is shown to be more amenable to genetic coding lthan other NN models. A hierarchical genetic coding scheme is developed for the bumptree and shown to have low redundancy, as well as being complete and closed with respect to the search space. When applied to optimising bumptree architectures for classification problems the GA discovers bumptrees which significantly out-perform those constructed using a standard algorithm. The fields of artificial life, control and robotics are identified as likely application areas for the evolutionary optimisation of NNs. An artificial life case-study is presented and discussed. Experiments are reported which show that the GA-bumptree is able to learn simulated pole balancing and car parking tasks using only limited environmental feedback. A simple modification of the fitness function allows the GA-bumptree to learn mappings which are multi-modal, such as robot arm inverse kinematics. The dynamics of the 'geographic speciation' selection model used by the GA-bumptree are investigated empirically and the convergence profile is introduced as an analytical tool. The relationships between the rate of genetic convergence and the phenomena of speciation, genetic drift and punctuated equilibrium arc discussed. The importance of genetic linkage to GA design is discussed and two new recombination operators arc introduced. The first, linkage mapped crossover (LMX) is shown to be a generalisation of existing crossover operators. LMX provides a new framework for incorporating prior knowledge into GAs.Its adaptive form, ALMX, is shown to be able to infer linkage relationships automatically during genetic search.

Neural networks for perpetual grouping

A number of researchers have investigated the application of neural networks to visual recognition, with much of the emphasis placed on exploiting the network's ability to generalise. However, despite the benefits of such an approach it is not at all obvious how networks can be developed which are capable of recognising objects subject to changes in rotation, translation and viewpoint. In this study, we suggest that a possible solution to this problem can be found by studying aspects of visual psychology and in particular, perceptual organisation. For example, it appears that grouping together lines based upon perceptually significant features can facilitate viewpoint independent recognition. The work presented here identifies simple grouping measures based on parallelism and connectivity and shows how it is possible to train multi-layer perceptrons (MLPs) to detect and determine the perceptual significance of any group presented. In this way, it is shown how MLPs which are trained via backpropagation to perform individual grouping tasks, can be brought together into a novel, large scale network capable of determining the perceptual significance of the whole input pattern. Finally the applicability of such significance values for recognition is investigated and results indicate that both the NILP and the Kohonen Feature Map can be trained to recognise simple shapes described in terms of perceptual significances. This study has also provided an opportunity to investigate aspects of the backpropagation algorithm, particularly the ability to generalise. In this study we report the results of various generalisation tests. In applying the backpropagation algorithm to certain problems, we found that there was a deficiency in performance with the standard learning algorithm. An improvement in performance could however, be obtained when suitable modifications were made to the algorithm. The modifications and consequent results are reported here.

A constructive learning algorithm based on back-propagation

There are been a resurgence of interest in the neural networks field in recent years, provoked in part by the discovery of the properties of multi-layer networks. This interest has in turn raised questions about the possibility of making neural network behaviour more adaptive by automating some of the processes involved. Prior to these particular questions, the process of determining the parameters and network architecture required to solve a given problem had been a time consuming activity. A number of researchers have attempted to address these issues by automating these processes, concentrating in particular on the dynamic selection of an appropriate network architecture.The work presented here specifically explores the area of automatic architecture selection; it focuses upon the design and implementation of a dynamic algorithm based on the Back-Propagation learning algorithm. The algorithm constructs a single hidden layer as the learning process proceeds using individual pattern error as the basis of unit insertion. This algorithm is applied to several problems of differing type and complexity and is found to produce near minimal architectures that are shown to have a high level of generalisation ability.

Neural network enhanced self tuning adaptive control application for non-linear control of dynamic systems

The main theme of research of this project concerns the study of neutral networks to control uncertain and non-linear control systems. This involves the control of continuous time, discrete time, hybrid and stochastic systems with input, state or output constraints by ensuring good performances. A great part of this project is devoted to the opening of frontiers between several mathematical and engineering approaches in order to tackle complex but very common non-linear control problems. The objectives are: 1. Design and develop procedures for neutral network enhanced self-tuning adaptive non-linear control systems; 2. To design, as a general procedure, neural network generalised minimum variance self-tuning controller for non-linear dynamic plants (Integration of neural network mapping with generalised minimum variance self-tuning controller strategies); 3. To develop a software package to evaluate control system performances using Matlab, Simulink and Neural Network toolbox. An adaptive control algorithm utilising a recurrent network as a model of a partial unknown non-linear plant with unmeasurable state is proposed. Appropriately, it appears that structured recurrent neural networks can provide conveniently parameterised dynamic models for many non-linear systems for use in adaptive control. Properties of static neural networks, which enabled successful design of stable adaptive control in the state feedback case, are also identified. A survey of the existing results is presented which puts them in a systematic framework showing their relation to classical self-tuning adaptive control application of neural control to a SISO/MIMO control. Simulation results demonstrate that the self-tuning design methods may be practically applicable to a reasonably large class of unknown linear and non-linear dynamic control systems.

A novel adaptive weight selection algorithm for multi-objective multi-agent reinforcement learning

To solve multi-objective problems, multiple reward signals are often scalarized into a single value and further processed using established single-objective problem solving techniques. While the field of multi-objective optimization has made many advances in applying scalarization techniques to obtain good solution trade-offs, the utility of applying these techniques in the multi-objective multi-agent learning domain has not yet been thoroughly investigated. Agents learn the value of their decisions by linearly scalarizing their reward signals at the local level, while acceptable system wide behaviour results. However, the non-linear relationship between weighting parameters of the scalarization function and the learned policy makes the discovery of system wide trade-offs time consuming. Our first contribution is a thorough analysis of well known scalarization schemes within the multi-objective multi-agent reinforcement learning setup. The analysed approaches intelligently explore the weight-space in order to find a wider range of system trade-offs. In our second contribution, we propose a novel adaptive weight algorithm which interacts with the underlying local multi-objective solvers and allows for a better coverage of the Pareto front. Our third contribution is the experimental validation of our approach by learning bi-objective policies in self-organising smart camera networks. We note that our algorithm (i) explores the objective space faster on many problem instances, (ii) obtained solutions that exhibit a larger hypervolume, while (iii) acquiring a greater spread in the objective space.

Performance of the Bayesian online algorithm for the perceptron

In this letter, we derive continuum equations for the generalization error of the Bayesian online algorithm (BOnA) for the one-layer perceptron with a spherical covariance matrix using the Rosenblatt potential and show, by numerical calculations, that the asymptotic performance of the algorithm is the same as the one for the optimal algorithm found by means of variational methods with the added advantage that the BOnA does not use any inaccessible information during learning. © 2007 IEEE.

Growing Neural Networks Using Nonconventional Activation Functions

In the paper, an ontogenic artificial neural network (ANNs) is proposed. The network uses orthogonal activation functions that allow significant reducing of computational complexity. Another advantage is numerical stability, because the system of activation functions is linearly independent by definition. A learning procedure for proposed ANN with guaranteed convergence to the global minimum of error function in the parameter space is developed. An algorithm for structure network structure adaptation is proposed. The algorithm allows adding or deleting a node in real-time without retraining of the network. Simulation results confirm the efficiency of the proposed approach.