| 和积网络的性质分析及其有效性验证算法 |
| |
| 作者姓名： | 刘洋 罗晨希 罗铁坚 |
| |
| 作者单位： | 1.中国科学院大学计算机与控制学院, 北京 101408;2.中国科学院软件所, 北京 100080 |
| |
| 基金项目： | 中国科学院仪器共享设备管理系统（Y42901VED2）资助 |
| |
| 摘 要： | 和积网络（sum-product networks，SPN）是一种在多层网络中进行快速推理的深度概率图模型，在人工智能领域有广泛应用前景。SPN的有效性即它可用来正确表示概率分布，使得SPN可以表示一些图模型的配分函数和所有的边缘分布。由于只有部分SPN是有效的，快速判断SPN的有效性很有必要。针对SPN理论体系中的有效性验证问题，讨论并揭示SPN内部结构性质，提出验证SPN有效性的两个算法，并给出算法的正确性证明及其复杂度。还通过给出一种新的SPN中生成树个数的计算方法来验证SPN有效性算法的可靠性。
|
| 关 键 词： | 深度学习 概率图模型 和积网络 有效性 |
| 收稿时间： | 2018-04-23 |
| 修稿时间： | 2018-07-18 |
Property analysis and validity verification algorithms of sum-product network |
| |
| Authors: | LIU Yang LUO Chenxi LUO Tiejian |
| |
| Institution: | 1.School of Computer and Control, University of Chinese Academy of Sciences, Beijing 101408, China;2.Institute of Software, Chinese Academy of Sciences, Beijing 100080, China |
| |
| Abstract: | Sum-product networks (SPN) is a deep probabilistic graphical model which has the characteristic of fast inference in multilayer networks, and it has wide application prospect in the field of artificial intelligence. The validity of SPN is that it can be used to represent the probability distribution correctly so that SPN can be used to represent the distribution functions of some graph models and all the marginal distributions. Since SPN is not always valid, it is necessary to determine the effectiveness of SPN quickly. In this paper, we consider the problem of validity verification in the SPN theoretical system, reveal the internal structure properties of SPN, and propose two algorithms for verifying the validity of SPN. The correctness proofs and the complexity of the proposed algorithms are given. We also verify the reliability of the proposed algorithms by giving a new method of calculating the number of generation trees in SPN. |
| |
| Keywords: | deep learning probabilistic graphical models sum-product networks validity |
|
| 点击此处可从《》浏览原始摘要信息 |
| 点击此处可从《》下载免费的PDF全文 |
