# Characterizing Multivariate Information Flows

@article{Hidaka2012CharacterizingMI, title={Characterizing Multivariate Information Flows}, author={Shohei Hidaka}, journal={ArXiv}, year={2012}, volume={abs/1212.5449} }

One of the crucial steps in scientiﬁc studies is to specify de pen-dent relationships among factors in a system of interest. Givenlittle knowledge of a system, can we characterize the underlyingdependent relationships through observation of its temporal be-haviors? In multivariate systems, there are potentially many pos-sible dependentstructures confusable with each other, and it maycause false detection of illusory dependency between unrelatedfactors. The present study proposes a new… Expand

#### References

SHOWING 1-10 OF 48 REFERENCES

CAUSALITY, FEEDBACK AND DIRECTED INFORMATION

- Mathematics
- 1990

It is shown that the "usual definition" of a discrete memoryless channel (DMC) in fact prohibits the use of feedback. The difficulty stems from the confusion of causality and statistical dependence.… Expand

Causality detected by transfer entropy leads acquisition of joint attention

- Computer Science
- 2007 IEEE 6th International Conference on Development and Learning
- 2007

In the computer simulation of human-robot interaction, which pair of perceptions and actions are selected as the causal pair are examined and it is shown that the selected pairs can be used to learn a sensorimotor map for achieving joint attention. Expand

The Multiinformation Function as a Tool for Measuring Stochastic Dependence

- Computer Science, Mathematics
- Learning in Graphical Models
- 1998

It is argued that the corresponding multiinformation function is a useful tool for problems concerning stochastic (conditional) dependence and independence (at least in the discrete case). Expand

Analyzing multimodal time series as dynamical systems

- Computer Science
- ICMI-MLMI '10
- 2010

A novel approach to discovering latent structures from multimodal time series based on the concept of generating partition which is the theoretically best symbolization of time series maximizing the information of the underlying original continuous dynamical system. Expand

Uncertainty and structure as psychological concepts

- Psychology
- 1975

It was a misfortune of psychology that it lacked a tradition of dealing with rigorous mathematical theories when psychologists were first attracted by information theory. Applications were made with… Expand

Spatio-Temporal Symbolization of Multidimensional Time Series

- Computer Science
- 2010 IEEE International Conference on Data Mining Workshops
- 2010

Probabilistic symbolic sequences derived from the symbolization method can be used in various supervised and unsupervised data-mining tasks and the new algorithm outperforms its alternative approaches. Expand

An approximation to the distribution of finite sample size mutual information estimates

- Mathematics, Computer Science
- IEEE International Conference on Communications, 2005. ICC 2005. 2005
- 2005

The distribution of mutual information between two discrete random variables is approximated by means of a second-order Taylor series expansion and conditional MI between conditionally independent variables and MI between (weakly) dependent random variables are derived. Expand

Approximating discrete probability distributions with causal dependence trees

- Mathematics, Computer Science
- 2010 International Symposium On Information Theory & Its Applications
- 2010

It is demonstrated that the minimum divergence approximation is the directed tree with maximum sum of directed informations, and a low-complexity minimum weight directed spanning tree, or arborescence, algorithm is specified to find the optimal tree. Expand

Complex network measures of brain connectivity: Uses and interpretations

- Computer Science, Medicine
- NeuroImage
- 2010

Construction of brain networks from connectivity data is discussed and the most commonly used network measures of structural and functional connectivity are described, which variously detect functional integration and segregation, quantify centrality of individual brain regions or pathways, and test resilience of networks to insult. Expand

Nonlinear Time Series Analysis

- Mathematics
- 1995

Abstract : This thesis applies neural network feature selection techniques to multivariate time series data to improve prediction of a target time series. Two approaches to feature selection are… Expand