Information-theoretic framework for unsupervised activity classification

被引：15

作者：

Kaplan, Frederic

Hafner, Verena V.

机构：

[1] Sony Corp, CSL Paris, F-75005 Paris, France

[2] Tech Univ Berlin, Fak Elektrotech & Informat, DAI Labor, D-10587 Berlin, Germany

来源：

ADVANCED ROBOTICS | 2006年 / 20卷 / 10期

基金：

欧盟地平线“2020”;

关键词：

activity classification; information metrics; unsupervised clustering;

D O I：

10.1163/156855306778522514

中图分类号：

TP24 [机器人技术];

学科分类号：

080202 ; 1405 ;

摘要：

This article presents a mathematical framework based on information theory to compare multivariate sensory streams. Central to this approach is the notion of configuration: a set of distances between information sources, statistically evaluated for a given time span. As information distances capture simultaneously effects of physical closeness, intermodality, functional relationship and external couplings, a configuration can be interpreted as a signature for specific patterns of activity. This provides ways for comparing activity sequences by viewing them as points in an activity space. Results of experiments with an autonomous robot illustrate how this framework can be used to perform unsupervised activity classification.

引用

页码：1087 / 1103

页数：17

共 50 条

[21] Information-theoretic feature selection for the classification of hysteresis curves
Gomez-Verdejo, Vanessa
Verleysen, Michel
Fleury, Jerome
[J]. COMPUTATIONAL AND AMBIENT INTELLIGENCE, 2007, 4507 : 522 - +
[22] Information-Theoretic Transfer Learning Framework for Bayesian Optimisation
Ramachandran, Anil
Gupta, Sunil
Rana, Santu
Venkatesh, Svetha
[J]. MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2018, PT II, 2019, 11052 : 827 - 842
[23] An information-theoretic framework for understanding saccadic eye movements
Lee, TS
Yu, SX
[J]. ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 12, 2000, 12 : 834 - 840
[24] An information-theoretic framework for process structure and data mining
Chiaravalloti, Antonio D.
Greco, Gianluigi
Guzzo, Antonella
Pontieri, Luigi
[J]. DATA WAREHOUSING AND KNOWLEDGE DISCOVERY, PROCEEDINGS, 2006, 4081 : 248 - 259
[25] An Automated Theorem Proving Framework for Information-Theoretic Results
Li, Cheuk Ting
[J]. 2021 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY (ISIT), 2021, : 2750 - 2755
[26] An Information-Theoretic Framework for Unifying Active Learning Problems
Nguyen, Quoc Phong
Low, Bryan Kian Hsiang
Jaillet, Patrick
[J]. THIRTY-FIFTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THIRTY-THIRD CONFERENCE ON INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE AND THE ELEVENTH SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2021, 35 : 9126 - 9134
[27] Information-theoretic feature selection algorithms for text classification
Novovicová, J
Malík, A
[J]. PROCEEDINGS OF THE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), VOLS 1-5, 2005, : 3272 - 3277
[28] Approaches to Information-Theoretic Analysis of Neural Activity
Victor J.D.
[J]. Biological Theory, 2006, 1 (3) : 302 - 316
[29] INFORMATION-THEORETIC ANALYSIS OF RANDOMIZATION AND SPONTANEOUS ACTIVITY
SHAFFER, JI
BARONE, NC
SLAK, S
[J]. PERCEPTUAL AND MOTOR SKILLS, 1984, 58 (01) : 55 - 60
[30] A unifying information-theoretic framework for independent component analysis
Lee, TW
Girolami, M
Bell, AJ
Sejnowski, TJ
[J]. COMPUTERS & MATHEMATICS WITH APPLICATIONS, 2000, 39 (11) : 1 - 21

← 1 2 3 4 5 →