arrow
返回

Active recursive Bayesian inference using Renyi information measures

delete2022-02-01
delete0
delete
OA
AI
Y
Yeganeh M. Marghi *
A
Aziz Koçanaoğulları
M
Murat Akçakaya
D
Deniz Erdoğmuş
DOI:10.1016/j.patrec.2022.01.009delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Recursive Bayesian inference (RBI) framework provides optimal Bayesian latent variable estimates in real -time settings with streaming noisy observations. Active RBI attempts to effectively select queries that lead to more informative observations to reduce uncertainty until the process is stopped at a certain confidence level. However, the mismatch between querying objective and stopping criterion creates the conundrum of improving the performance of one objective at the expense of deteriorating the other. Moreover, conventional active querying methods stagger in the presence of misleading prior information. Inspired by information theoretic approaches, we propose an active RBI framework where query and stop-ping criterion are jointly selected through a unified objective based on Renyi information measures. The proposed unified formulation enables us to jointly enhance speed and accuracy in the RBI process. We theoretically demonstrate that the proposed objective encourages exploration in the presence of mislead-ing prior. Furthermore, we motivate our framework by proving a geometrical representation for active querying and decision making on a probability simplex. We provide empirical and experimental studies on two applications including restaurant recommendation and brain-computer interface (BCI) typing sys-tems and demonstrate that our method outperforms comparable approaches by improving both speed and accuracy.(c) 2022 Elsevier B.V. All rights reserved.
Keyword:
Active learning
Recursive state estimation
Bayesian inference
Renyi entropy
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Pattern Recognition Letters 封面图
Pattern Recognition Letters
IF:
3.3
论文数:
7.9K
被引数:
1.6W

机构

N
Northeastern University
学者数:
2.5W
论文数: 1.6W
被引数: 3.0W
P
pennsylvania commonwealth system of higher education (pcshe)
学者数:
12.9W
论文数: 11.7W
被引数: 177
引用论文

引用论文

Brain-Computer Interface With Language Model-Electroencephalography Fusion for Locked-In Syndrome
err2013-12-26
err61
errOAAI
errOken, Barry S.; Orhan, Umut; Roark, Brian; Erdogmus, Deniz; Fowler, Andrew; Mooney, Aimee; Peters, Betts; Miller, Meghan; Fried-Oken, Melanie B.
err分享
err收藏
err分享
err收藏
err分享
err收藏
err分享
err收藏
学者 查看更多内容