Vecchia approximations provide the best accuracy-runtime trade-off among the tested scalable Gaussian process approximations for spatial data.
19 Appendix A.1 Additional figures Figure 16: Average negative log-likelihood on simulated data sets with a practical range of 0.2 and a sample sizeN= 100,000
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.CO 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
An accuracy-runtime trade-off comparison of scalable Gaussian process approximations for spatial data
Vecchia approximations provide the best accuracy-runtime trade-off among the tested scalable Gaussian process approximations for spatial data.