We explain the different approaches to fusing neural networks and symbolic processing aspects. Four categories are distinguished namely fusion, transformation, combination and associative systems. A brief description ...
详细信息
We explain the different approaches to fusing neural networks and symbolic processing aspects. Four categories are distinguished namely fusion, transformation, combination and associative systems. A brief description of each is given. We also briefly describe the GEMSES architecture as an example of an associative system which has been built from philosophical, cognitive, practical viewpoints.
Despite the performance potential of parallel systems, several factors have hindered their widespread adoption. Of these, performance variability is among the most significant. Data parallel languages, which facilitat...
详细信息
ISBN:
(纸本)9780818685910
Despite the performance potential of parallel systems, several factors have hindered their widespread adoption. Of these, performance variability is among the most significant. Data parallel languages, which facilitate the programming of those systems, increase the semantic distance between the program's source code and its observable performance, thus aggravating the optimization problem. In this paper, we present a new methodology to automatically predict the performance scalability of data parallel applications on multicomputers. Our technique represents the execution time of a program as a symbolic expression that includes the number of processors (P), problem size (N), and other system-dependent parameters. This methodology is strongly based on information collected at compile time. By extending an existing data parallel compiler (Fortran D95), we derive during compilation, a symbolic cost model that represents the expected cost of each high-level code section and, inductively, of the complete program.
Bayesian treatments of learning in neural networks are typically based either on local Gaussian approximations to a mode of the posterior weight distribution, or on Markov chain Monte Carlo simulations. A third approa...
详细信息
ISBN:
(纸本)0262100762
Bayesian treatments of learning in neural networks are typically based either on local Gaussian approximations to a mode of the posterior weight distribution, or on Markov chain Monte Carlo simulations. A third approach, called ensemble learning, was introduced by Hinton and van Camp (1993). It aims to approximate the posterior distribution by minimizing the Kullback-Leibler divergence between the true posterior and a parametric approximating distribution. However, the derivation of a deterministic algorithm relied on the use of a Gaussian approximating distribution with a diagonal covariance matrix and so was unable to capture the posterior correlations between parameters. In this paper, we show how the ensemble learning approach can be extended to full- covariance Gaussian distributions while remaining computationally tractable. We also extend the framework to deal with hyperparam- eters, leading to a simple re-estimation procedure. Initial results from a standard benchmark problem are encouraging.
The purpose of this paper is to examine the problem of controlling a linear discrete-time system subject to input saturation in order to have its output track (or reject) a family of reference (or disturbance) signals...
详细信息
In this paper we outline an intelligent hybrid multiagent architecture for engineering complex systems. The hybrid multi-agent architecture is described at the task structure level. The architecture has been successfu...
详细信息
ISBN:
(纸本)0780341236
In this paper we outline an intelligent hybrid multiagent architecture for engineering complex systems. The hybrid multi-agent architecture is described at the task structure level. The architecture has been successfully applied in a real time alarm processing and fault diagnosis system in a power system control centre with good performance results.
For neural networks with a wide class of weight-priors, it can be shown that in the limit of an infinite number of hidden units the prior over functions tends to a Gaussian process. In this paper analytic forms are de...
详细信息
ISBN:
(纸本)0262100657
For neural networks with a wide class of weight-priors, it can be shown that in the limit of an infinite number of hidden units the prior over functions tends to a Gaussian process. In this paper analytic forms are derived for the covariance function of the Gaussian processes corresponding to networks with sigmoidal and Gaussian hidden units. This allows predictions to be made efficiently using networks with an infinite number of hidden units, and shows that, somewhat paradoxically, it may be easier to compute with infinite networks than finite ones.
The full Bayesian method for applying neural networks to a prediction problem is to set up the prior/hyperprior structure for the net and then perform the necessary integrals. However, these integrals are not tractabl...
详细信息
ISBN:
(纸本)0262100657
The full Bayesian method for applying neural networks to a prediction problem is to set up the prior/hyperprior structure for the net and then perform the necessary integrals. However, these integrals are not tractable analytically, and Markov Chain Monte Carlo (MCMC) methods are slow, especially if the parameter space is high-dimensional. Using Gaussian processes we can approximate the weight space integral analytically, so that only a small number of hyperparameters need be integrated over by MCMC methods. We have applied this idea to classification problems, obtaining excellent results on the real-world problems investigated so far.
Principal component analysis (PCA) is a ubiquitous technique for data analysis but one whose effective application is restricted by its global linear character. While global nonlinear variants of PCA have been propose...
Principal component analysis (PCA) is a ubiquitous technique for data analysis but one whose effective application is restricted by its global linear character. While global nonlinear variants of PCA have been proposed, an alternative paradigm is to capture data nonlinearity by a mixture of local PCA models. However, existing techniques are limited by the absence of a probabilistic formalism with an appropriate likelihood measure and so require an arbitrary choice of implementation strategy. This paper shows how PCA can be derived from a maximum-likelihood procedure, based on a specialisation of factor analysis. This is then extended to develop a well-defined mixture model of principal component analyzers, and an expectation-maximisation algorithm for estimating all the model parameters is given.
The more complex a network becomes, the more reliable and intelligent a network management system must be to consistently monitor the network and detect abnormal situations in a timely manner as they occur. Expert sys...
详细信息
The more complex a network becomes, the more reliable and intelligent a network management system must be to consistently monitor the network and detect abnormal situations in a timely manner as they occur. Expert system techniques have been widely accepted to create network management systems. Despite the fact that there are a number of network management systems, most of them deal only with problems at the lower layers of the network hierarchy (the data link, or the network layer). The nature of problems at the application level significantly differ from those that occur at the lower levels. Lower layer problems are well understood while problems at the application level are complex, application dependent, and distinct from one another. Consequently, a network management system, in particular a fault management system, used at this level should be able to cope with these difficulties and dependencies. We propose a hybrid system which consists of neural network module and a rule based system for monitoring and diagnosing problems occur at the application level. The domain name system (DNS) was selected as a testbed application for the prototype system.
暂无评论