Abstract | ||
---|---|---|
Cross-validation (CV) is a technique for evaluating the ability of statistical models/learning systems based on a given data set. Despite its wide applicability, the rather heavy computational cost can prevent its use as the system size grows. To resolve this difficulty in the case of Bayesian linear regression, we develop a formula for evaluating the leave-one-out CV error approximately without actually performing CV. The usefulness of the developed formula is tested by statistical mechanical analysis for a synthetic model. This is confirmed by application to a real-world supernova data set as well. |
Year | DOI | Venue |
---|---|---|
2016 | 10.1109/ALLERTON.2016.7852286 | 2016 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton) |
Keywords | DocType | Volume |
cross-validation formula approximation,Bayesian linear regression,statistical models,learning systems,leave-one-out CV error,statistical mechanical analysis,synthetic model | Conference | abs/1610.07733 |
ISSN | ISBN | Citations |
2474-0195 | 978-1-5090-4551-8 | 0 |
PageRank | References | Authors |
0.34 | 6 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yoshiyuki Kabashima | 1 | 136 | 27.83 |
Tomoyuki Obuchi | 2 | 11 | 5.65 |
Makoto Uemura | 3 | 7 | 4.54 |