A Genetic Adaptive System for Image Understanding and Learning Research. Phase 1
Abstract
This report documents the efforts and results of initial phase research on machine learning directed at application for real time machine vision and automatic target recognition. The particular paradigm pursued is based on genetic algorithms and classifiers modeled on the summation of Mendelian genetic recombination, Darwinian selection and ecological notions of competition. This machine learning approach is strongly supported by sound statistical theory. A second thread of research was the development of massively parallel computing hardware based on the Geometric/Arithmetic Parallel Processor (GAPP). This machine has a large number of processors, each one bit wide with a full Arithmetic/Logic Unit (full adder) and with local memory per processor. The basic research hypothesis of the subject effort has been that GAPP contained sufficient hardware capability to provide a substrate for a Classifier and Genetic Algorithm system. The goal has been demonstrated by constructing and running the necessary software on the GAPP, its controller and its host. The resulting fusion of software and hardware is called a Genetic Algorithm/ Classifier Engine (GACE) in the same sense as a LISP engine or a database engine. The resulting quantum jump in performance should open doors both to application and to more interesting and relevant research. (SDW)
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1989
- Accession Number
- ADA214810
Entities
People
- Dean Z. Douthat
- Kevin W. Ross