| US 7,496,549 B2 | ||
| Matching pursuit approach to sparse Gaussian process regression | ||
| Sathiya Keerthi Selvaraj, South Pasadena, Calif. (US) | ||
| Assigned to Yahoo! Inc., Sunnyvale, Calif. (US) | ||
| Filed on Nov. 18, 2005, as Appl. No. 11/282,306. | ||
| Claims priority of provisional application 60/685660, filed on May 26, 2005. | ||
| Prior Publication US 2006/0271532 A1, Nov. 30, 2006 | ||
| Int. Cl. G06N 5/00 (2006.01) | ||
| U.S. Cl. 706—25 [706/45] | 34 Claims |
| 1. A computerized method for supervised learning by performing a Gaussian process to select and yield a set of search items
to a user, comprising:
receiving a set of training elements;
selecting a basis element from the set of training elements;
adding the basis element to a basis element set;
conducting an optimization test on the basis element set with the selected basis element to produce a selection score;
determining whether the selection score indicates an improvement in optimization for the basis element set;
discarding the basis element if the selection score does not indicate an improvement; and
retaining the basis element in the basis element set if the selection score indicates an improvement; and
providing the basis element set to the Gaussian process for supervised learning to select and yield a set of search items
to a user.
|