Please use this identifier to cite or link to this item:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/9011
Title: | 應用截斷牛頓法於條件隨機場 Newton Methods for Conditional Random Fields |
Authors: | Peng-Jen Chen 陳鵬仁 |
Advisor: | 林智仁(Chih-Jen Lin) |
Keyword: | 共軛梯度法,信賴區間牛頓法,最大熵值法,條件隨機場, conjugate gradient methods,trust region Newton methods,maximum entropy,conditional random fields, |
Publication Year : | 2009 |
Degree: | 碩士 |
Abstract: | 條件隨機場是一個適合用來標記序列性資料的模組。由於考慮序列中所有可能的標籤組合,條件隨機場在學習及預測階段都非常耗時。牛頓法在最佳化的最後階段具有較快的收斂性質,因此我們採用牛頓法來解條件隨機場。海森矩陣向量乘積是整個計算過程中最耗時的部份。本篇論文提出一個新的動態規劃技巧,可以在多項式時間複雜度內完成海森矩陣向量乘積。 Conditional Random Fields (CRFs) is a useful technique to label sequential data. Due to considering all label combinations of a sequence, CRFs' training and testing are time consuming. In this work, we consider a Newton method for training CRFs because of its possible fast final convergence. The computational bottleneck is on the Hessian-vector product. We propose a novel dynamic programming technique to calculate it in polynomial time. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/9011 |
Fulltext Rights: | 同意授權(全球公開) |
Appears in Collections: | 資訊工程學系 |
Files in This Item:
File | Size | Format | |
---|---|---|---|
ntu-98-1.pdf | 1.4 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.