Mathematically Inspired Environment Classification and Identification: Simplicial Complex Hierarchy

Abstract

We provide a mathematical foundation for a biologically inspired hierarchical identification and classification system. We describe lines detected by the viual cortex in terms of natural basis elements, upon which simplicial complexes are erected to describe simple objects. Simple objects are basis elements in the next level of the hierarchy, and objects are descirbed simply in terms of them. This is continued through several levels of the hierarchy, and each level identifies somewhat more complex objects onto a manifold the previous level. Projections of the identified objects onto a manifold provides a natural and useful representation of the system's environment.

Meeting Name

Artificial Neural Networks in Engineering Conference, ANNIE 2004 (2004: Nov. 7-10, St. Louis, MO)

Department(s)

Mathematics and Statistics

Second Department

Electrical and Computer Engineering

Keywords and Phrases

Neural networks; Artificial life; Complex systems; Fuzzy logic; Programming

International Standard Book Number (ISBN)

0-7918-0228-0

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2004 American Society of Mechanical Engineers (ASME), All rights reserved.

Publication Date

10 Nov 2004

This document is currently not available here.

Share

 
COinS