期刊
IEEE SIGNAL PROCESSING LETTERS
卷 10, 期 4, 页码 115-118出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LSP.2003.809034
关键词
dependence tree models; hidden Markov models; Kullback-Leibler distance
We present a fast algorithm to approximate the Kullback-Leibler distance (KLD) between two dependence tree models. The algorithm uses the upward (or forward) procedure to compute an upper bound for the KLD. For hidden Markov models, this algorithm is reduced to a simple expression. Numerical experiments show that for a similar accuracy, the proposed algorithm offers a saving of hundreds of times in computational complexity compared to the commonly used Monte Carlo method. This makes the proposed algorithm important for real-time applications, such as image retrieval.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据