Name
Abstract | ||
|---|---|---|
Modeling of conditional quantiles requires specification of the quantile being estimated and can thus be viewed as a parameterized predictive modeling problem. Quantile loss is typically used, and it is indeed parameterized by a quantile parameter. In this paper we show how to follow the path of cross validated solutions to regularized kernel quantile regression. Even though the bi-level optimization problem we encounter for every quantile is non-convex, the manner in which the optimal cross-validated solution evolves with the parameter of the loss function allows tracking of this solution. We prove this property, construct the resulting algorithm, and demonstrate it on data. This algorithm allows us to efficiently solve the whole family of bi-level problems. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1145/1390156.1390262 | Journal of Machine Learning Research |
Keywords | DocType | Volume |
bi-level optimization problem,cross validated solution,quantile loss,conditional quantiles,resulting algorithm,regularized kernel quantile regression,regularized optimization problem,kernel quantile regression,optimal cross-validated solution evolves,bi-level problem,parameterized loss function,support vector regression,loss function,parameterized predictive modeling problem,bi-level path,alternative in-sample model selection,quantile parameter | Conference | 10, |
ISSN | Citations | PageRank |
1532-4435 | 6 | 0.60 |
References | Authors | |
13 | 1 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Saharon Rosset | 1 | 1087 | 105.33 |