发明名称 |
TRAINING TREE TRANSDUCERS |
摘要 |
Training using tree transducers is described. Given sample input/output pairs as training (100, 110), and given a set of tree transducer rules (120), the information is combined to yield locally optimal weights for those rules (140). This combination is carried out by building a weighted derivation forest for each input/output pair and applying counting methods to those forests (130). |
申请公布号 |
WO2005089340(A3) |
申请公布日期 |
2008.01.10 |
申请号 |
WO2005US08648 |
申请日期 |
2005.03.15 |
申请人 |
UNIVERSITY OF SOUTHERN CALIFORNIA;KNIGHT, KEVIN;GRAEHL, JONATHAN |
发明人 |
KNIGHT, KEVIN;GRAEHL, JONATHAN |
分类号 |
G06F17/28;G06F17/27 |
主分类号 |
G06F17/28 |
代理机构 |
|
代理人 |
|
主权项 |
|
地址 |
|