A dependency-based framework of combining multiple experts for the recognition of unconstrained handwritten numerals

Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149) Pub Date : 1999-06-23 DOI:10.1109/CVPR.1999.784619

H. Kang, Seong-Whan Lee

{"title":"A dependency-based framework of combining multiple experts for the recognition of unconstrained handwritten numerals","authors":"H. Kang, Seong-Whan Lee","doi":"10.1109/CVPR.1999.784619","DOIUrl":null,"url":null,"abstract":"Although Behavior-Knowledge Space (BKS) method does not need any assumptions in combining multiple experts, it should build theoretically exponential storage spaces for storing and managing jointly observed K decisions from K experts. That is, combining K experts needs a (K+1)st-order probability distribution. However, it is well known that the distribution becomes unmanageable in storing and estimating, even for a small K. In order to overcome such weakness, it would be attractive to decompose the distribution into a number of component distributions and to approximate the distribution with a product of the component distributions. One of such previous works is to apply a conditional independence assumption to the distribution. Another work is to approximate the distribution with a product of only first-order tree dependencies or second-order distributions. In this paper, a dependency-based framework is proposed to optimality approximate a probability distribution with a product set of dth-order dependencies where 1<d<K, and to combine multiple experts based on the product set using the Bayesian formalism. This framework was experimented and evaluated with a standardized CEN-PARIMl data base.","PeriodicalId":20644,"journal":{"name":"Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"1999-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"9","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CVPR.1999.784619","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 9

Abstract

Although Behavior-Knowledge Space (BKS) method does not need any assumptions in combining multiple experts, it should build theoretically exponential storage spaces for storing and managing jointly observed K decisions from K experts. That is, combining K experts needs a (K+1)st-order probability distribution. However, it is well known that the distribution becomes unmanageable in storing and estimating, even for a small K. In order to overcome such weakness, it would be attractive to decompose the distribution into a number of component distributions and to approximate the distribution with a product of the component distributions. One of such previous works is to apply a conditional independence assumption to the distribution. Another work is to approximate the distribution with a product of only first-order tree dependencies or second-order distributions. In this paper, a dependency-based framework is proposed to optimality approximate a probability distribution with a product set of dth-order dependencies where 1

查看原文本刊更多论文

结合多个专家的基于依赖的框架，用于无约束手写数字的识别

虽然行为知识空间(BKS)方法在组合多个专家时不需要任何假设，但它应该建立理论上的指数存储空间来存储和管理K个专家共同观察到的K个决策。也就是说，组合K个专家需要一个(K+1)个一阶概率分布。然而，众所周知，即使对于一个很小的k，分布在存储和估计方面也变得难以管理。为了克服这种弱点，将分布分解为许多分量分布，并用分量分布的乘积来近似分布是很有吸引力的。以前的工作之一是对分布应用条件独立假设。另一项工作是用一阶树依赖关系或二阶分布的乘积来近似分布。本文提出了一种基于依赖关系的框架，用于最优逼近具有1

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Proceedings. 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No PR00149)

自引率

0.00%

发文量

京公网安备 11010802042870号

Book学术文献互助群
群号：481959085