We propose a Bayesian framework for regression problems, which covers areas usually dealt with by function approximation. An online learning algorithm is derived which solves regression problems with a Kalman filter. ...
详细信息
We propose a Bayesian framework for regression problems, which covers areas usually dealt with by function approximation. An online learning algorithm is derived which solves regression problems with a Kalman filter. its solution always improves with increasing model complexity, without the risk of over-fitting. In the infinite dimension limit it approaches the hue Bayesian posterior. The issues of prior selection and over-fitting are also discussed, showing that some of the commonly held beliefs are misleading. The practical implementation is summarised. Simulations using 13 popular publicly available data sets are used to demonstrate the method and highlight important issues concerning the choice of priors.
It is known theoretically that an algorithm cannot be good for an arbitrary prior. We show that in practical terms this also applies to the technique of ''cross-validation,'' which has been widely rega...
It is known theoretically that an algorithm cannot be good for an arbitrary prior. We show that in practical terms this also applies to the technique of ''cross-validation,'' which has been widely regarded as defying this general rule. Numerical examples are analyzed in detail. Their implications to researches on learning algorithms are discussed.
Most conventional techniques for estimating conditional probability densities are inappropriate for applications involving periodic variables. In this paper we introduce three related techniques for tackling such prob...
Most conventional techniques for estimating conditional probability densities are inappropriate for applications involving periodic variables. In this paper we introduce three related techniques for tackling such problems, and investigate their performance using synthetic data. We then apply these techniques to the problem of extracting the distribution of wind vector directions from radar scatterometer data gathered by a remote-sensing satellite.
The paper describes a software engineering subject on system description techniques (SDT). This course explains the system description techniques (SDTs) for describing both the models that arise from analysis as well ...
详细信息
In this paper we introduce an abstract data type for the distributed representation and efficient handling of sparse grids on parallel architectures. The new data layout, implemented by means of the PVM message passin...
详细信息
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...
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...
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.
A discussion group met during the conference to foster discussion of some social issues in software engineering (SE). Three initial questions were posed to stimulate and focus the deliberations. First, to what extent ...
详细信息
A discussion group met during the conference to foster discussion of some social issues in software engineering (SE). Three initial questions were posed to stimulate and focus the deliberations. First, to what extent should ethics be taught in SE courses, and at what level. Second, are issues such as privacy and confidentiality adequately covered in current SE courses? Thirdly, is ethical self-assessment by students an effective learning method?.
A low gain design for linear discrete-time systems subject to input saturation was recently developed in (Lin and Saberi 1995a, Lin et al. 1995a, Mantri et al. 1995) to solve both semi-global stabilization and semi-gl...
详细信息
A low gain design for linear discrete-time systems subject to input saturation was recently developed in (Lin and Saberi 1995a, Lin et al. 1995a, Mantri et al. 1995) to solve both semi-global stabilization and semi-global output regulation problems. This paper proposes an improvement to the low gain design and determines controllers with the new design that achieve semi-global output regulation. The improvement is reflected in better utilization of available control capacity and consequently better closed-loop performance.
In this paper, we show the relevance of fuzzy systems in an integrated symbolic-subsymbolic architecture, GENUES (generic neuro-expert system), for information processing. Fuzzy logic is used for modelling knowledge i...
详细信息
In this paper, we show the relevance of fuzzy systems in an integrated symbolic-subsymbolic architecture, GENUES (generic neuro-expert system), for information processing. Fuzzy logic is used for modelling knowledge in different phases of the GENUES architecture. As an illustration we show the application of fuzzy logic in the decision phase and post-processing phase of GENUES for power system fault diagnosis.
暂无评论