Name
Abstract | ||
|---|---|---|
Gaussian Processes (GPs) are Bayesian non-parametric models that achieve state-of-the-art performance in regression tasks. To allow for analytical tractability, noise power is usually considered constant in these models, which is unrealistic for many real world problems. In this work we consider a GP model with heteroscedastic (i.e., input dependent) noise power, and then, use Expectation Propagation (EP) to perform approximate inference on it. The proposed EP approach is much faster than Markov Chain Monte Carlo and more accurate than competing methods of similar computational cost. This superiority is illustrated in several experiments with synthetic and real-world data. |
Year | DOI | Venue |
|---|---|---|
2011 | 10.1109/MLSP.2011.6064576 | 2011 IEEE INTERNATIONAL WORKSHOP ON MACHINE LEARNING FOR SIGNAL PROCESSING (MLSP) |
Keywords | Field | DocType |
markov chain monte carlo,gaussian process,gaussian process regression,parametric model | Kriging,Noise power,Gaussian random field,Markov chain Monte Carlo,Pattern recognition,Computer science,Approximate inference,Artificial intelligence,Gaussian process,Expectation propagation,Bayesian probability | Conference |
ISSN | Citations | PageRank |
2161-0363 | 5 | 0.48 |
References | Authors | |
6 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Luis Muñoz-González | 1 | 82 | 8.48 |
| Miguel Lázaro-Gredilla | 2 | 383 | 26.46 |
| Aníbal R. Figueiras-Vidal | 3 | 467 | 38.03 |