About this document
Distributed Gaussian Process Regression by hadjiamine93 is a document available to read on EtoBox.
The document presents a simulation study comparing various Gaussian process regression techniques using distributed methods. It discusses four methods of data partitioning and their effects on posterior mean estimation, convergence rates, and computational efficiency. The results indicate that while distributed methods can yield comparable performance to non-distributed methods, they may also introduce challenges such as discontinuities and suboptimal contraction rates.
- Author
- hadjiamine93
- Language
- EN