Optimal representation of experimental data

No single state space reconstruction of experimental data is optimal for all scientific endeavors. This thesis is all about the coordinates used in reconstruction, the goals of reconstruction, and the criteria used to select optimal reconstructions. A learning algorithm is described, which when given a space of possible coordinates and an optimality criterion selects the best state space reconstruction. This method is applied to specific problems: modeling nonuniformly sampled data, computing the inferable dimension of a data set, and model-based control.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Physics

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2011

Author dc:creator

• Breeden, Joseph Lynn

Contributors dc:contributor

• Packard, Norman H.

• Physics, General
• Physics, Astronomy and Astrophysics

Statement dc:rights

• Copyright 1992 Breeden, Joseph Lynn

Language dc:language

eng

Identifier

AAI9236400
(UMI)AAI9236400

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/22671