CASIA OpenIR  > 精密感知与控制研究中心  > 人工智能与机器学习
大数据下的PAC-Bayesian学习理论综述
杨雪冰; 张文生; 杨阳
Source Publication山西大学学报(自然科学版)
2015
Volume38Issue:3Pages:413-419
Abstract概率近似正确(PAC)是研究“可学习”的理论框架。近年来,研究人员融合贝叶斯方法与不依赖分布的PAC性能度量提出了所谓的PAC-Bayesian学习理论。该理论因其对于任意概念空间任意测度的先验均能给出泛化误差界而在人工智能的不同领域的相关算法分析中得到广泛应用。本文综述了PAC-Bayesian学习理论的由来及其核心思想,进而结合大数据的特点,论述了PAC-Bayesian适合于大数据相关算法的理论分析。
Other AbstractThe theory of Probably Approximately Correct (PAC) is a framework for the study of learnable. In recent years, researchers combined Bayesian method with distribution-free PAC guarantees and proposed so-called PAC-Bayesian learning theory. This theory has been widely used in different fields of Artificial Intelligence to analyze related algorithms for the given generalization error bounds can apply to an arbitrary prior measure on an arbitrary concept space. This paper surveys the derivation of PAC-Bayesian learning theory and its core ideas. Further, considering the characteristics of big data, this paper discusses why PAC-Bayesian is useful for theoretical analysis of the related algorithms for big data.
KeywordPac-bayesian界 学习理论 贝叶斯学习 大数据
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/19677
Collection精密感知与控制研究中心_人工智能与机器学习
Corresponding Author张文生
Affiliation中国科学院自动化研究所
Recommended Citation
GB/T 7714
杨雪冰,张文生,杨阳. 大数据下的PAC-Bayesian学习理论综述[J]. 山西大学学报(自然科学版),2015,38(3):413-419.
APA 杨雪冰,张文生,&杨阳.(2015).大数据下的PAC-Bayesian学习理论综述.山西大学学报(自然科学版),38(3),413-419.
MLA 杨雪冰,et al."大数据下的PAC-Bayesian学习理论综述".山西大学学报(自然科学版) 38.3(2015):413-419.
Files in This Item: Download All
File Name/Size DocType Version Access License
大数据下的PAC-Bayesian学习理(748KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[杨雪冰]'s Articles
[张文生]'s Articles
[杨阳]'s Articles
Baidu academic
Similar articles in Baidu academic
[杨雪冰]'s Articles
[张文生]'s Articles
[杨阳]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杨雪冰]'s Articles
[张文生]'s Articles
[杨阳]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 大数据下的PAC-Bayesian学习理论.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.