Abstract | ||
---|---|---|
We propose a scalable multiple-output generalization of unscented and extended Gaussian processes. These algorithms have been designed to handle general likelihood models by linearizing them using a Taylor series or the Unscented Transform in a variational inference framework. We build upon random feature approximations of Gaussian process covariance functions and show that, on small-scale single-task problems, our methods can attain similar performance as the original algorithms while having less computational cost. We also evaluate our methods at a larger scale on MNIST and on a seismic inversion which is inherently a multi-task problem. |
Year | Venue | Field |
---|---|---|
2016 | ICML | Seismic inversion,Mathematical optimization,MNIST database,Inference,Computer science,Unscented transform,Gaussian process,Artificial intelligence,Machine learning,Taylor series,Covariance,Scalability |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
15 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Edwin V. Bonilla | 1 | 1008 | 53.32 |
Daniel M. Steinberg | 2 | 35 | 2.85 |
Alistair Reid | 3 | 0 | 0.34 |