Approximate Inference for Nonstationary Heteroscedastic Gaussian process Regression
This work addresses computational efficiency in Gaussian process regression for scenarios with input-dependent variances, offering a practical solution for machine learning applications, though it is incremental as it builds on existing inference techniques.
The paper tackles approximate inference for nonstationary heteroscedastic Gaussian process regression by developing an efficient method using expectation propagation to integrate over noise and signal variances, showing comparable results to Markov chain Monte Carlo with reduced computational burden.
This paper presents a novel approach for approximate integration over the uncertainty of noise and signal variances in Gaussian process (GP) regression. Our efficient and straightforward approach can also be applied to integration over input dependent noise variance (heteroscedasticity) and input dependent signal variance (nonstationarity) by setting independent GP priors for the noise and signal variances. We use expectation propagation (EP) for inference and compare results to Markov chain Monte Carlo in two simulated data sets and three empirical examples. The results show that EP produces comparable results with less computational burden.