RESEARCH COMMONS
LIBRARY

New Developments In Cognitive Optimization

ResearchCommons/Manakin Repository

New Developments In Cognitive Optimization

Show full item record

Title: New Developments In Cognitive Optimization
Author: Kandala, Parvati Aruna
Abstract: This work presents an approach to solve function optimization problems using cognitive optimization. A cognitive search algorithm which is developed mimics human cognition in structure and characteristics. This algorithm is developed based on the key aspects of a cognitive system, namely, the use of prior knowledge, communication, and self-organization. The processes have the knowledge of how to solve the optimization problems from traditional techniques such as bracketing, gradient search and interval-halving methods. This search algorithm shows that, by passing messages, the processes communicate, share information, and self-organize around the solution. A mat lab program is coded to perform the tests on difficult problems which are not solved by the traditional bracketing and gradient search techniques alone. This work will be applied on complex functions such as non-convex, multiple local optima and discontinuous functions. The results for the different cases considered explore the structure and characteristics of cognitive optimization, which provide guidelines for the development of general cognitive systems for smart devices.
URI: http://hdl.handle.net/10106/11083
Date: 2012-07-25

Files in this item

Files Size Format View
Kandala_uta_2502M_11588.pdf 542.6Kb PDF View/Open

This item appears in the following Collection(s)

Show full item record

Browse

My Account

Statistics

About Us