Gaussian Processes for Bayesian Classiication via Hybrid Monte Carlo
نویسندگان
چکیده
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 inte-grals 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 classiication problems, obtaining excellent results on the real-world problems investigated so far.
منابع مشابه
Gaussian Processes for Regression
The Bayesian analysis of neural networks is difficult because a simple prior over weights implies a complex prior distribution over functions . In this paper we investigate the use of Gaussian process priors over functions, which permit the predictive Bayesian analysis for fixed values of hyperparameters to be carried out exactly using matrix operations. Two methods, using optimization and aver...
متن کاملGaussian Processes for Bayesian Classification via Hybrid Monte Carlo
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 ...
متن کاملGaussian Processes to Speed up Hybrid Monte Carlo for Expensive Bayesian Integrals
Hybrid Monte Carlo (HMC) is often the method of choice for computing Bayesian integrals that are not analytically tractable. However the success of this method may require a very large number of evaluations of the (un-normalized) posterior and its partial derivatives. In situations where the posterior is computationally costly to evaluate, this may lead to an unacceptable computational load for...
متن کاملBayesian Regression and Classification Using Mixtures of Gaussian Processes
For a large data-set with groups of repeated measurements, a mixture model of Gaussian process priors is proposed for modelling the heterogeneity among the different replications. A hybrid Markov chain Monte Carlo (MCMC) algorithm is developed for the implementation of the model for regression and classification. The regression model and its implementation are illustrated by modelling observed ...
متن کاملHybrid Bayesian M-Estimation and Its Application in Calibration and Tuning for Complex Computer Models
Complex computer simulation is critical in biomedical research fields, e.g., the design of prosthetic devices. Tuning and calibration are processes for improving the representativeness of a computer simulation code to a physical phenomenon, which is the key to the success of a computer simulator. In this talk I will introduce a statistical methodology for simultaneously determining tuning and c...
متن کامل