Distributed Adaptation in Neuromime Networks
Report Number: AMRL TR 66-225
Author(s): Griffith, V. V., Bolen, G. H.
Corporate Author(s): Goodyear Aerospace Corporation
Laboratory: Aerospace Medical Research Laboratories
Date of Publication: 1967-01
Pages: 96
Contract: AF 33(615)-1891
DoD Project: 7232
DoD Task: 723203
Identifier: AD0652853
Abstract:
The report describes investigations of networks with adaptive ability distributed through them. It is thought that large-scale systems can be constructed of adaptive building blocks. These adaptive systems would be flexible in function, reliable and would resist severe damage characteristics of living creatures. Neuron models were tested by interconnecting them into various networks to perform simple control tasks. The test results were evaluated and the evaluation used to improve the theory and the neuron model. Two basic analysis methods were used to study neuromime networks: a sequential machine analysis and an optimal process method applying Pontryagin's maximum principle. The sequential analysis method proved unsatisfactory when applied to an attempted description of an adjustment rule for the neuron model. This difficulty led to an application of optimal processes. The application of Pontryagin's maximum principle to the analysis of the neuron model network described both optimum conditions for a system and criteria useful for developing the adjustment rule.
Provenance: RAF Centre of Aviation Medicine
Author(s): Griffith, V. V., Bolen, G. H.
Corporate Author(s): Goodyear Aerospace Corporation
Laboratory: Aerospace Medical Research Laboratories
Date of Publication: 1967-01
Pages: 96
Contract: AF 33(615)-1891
DoD Project: 7232
DoD Task: 723203
Identifier: AD0652853
Abstract:
The report describes investigations of networks with adaptive ability distributed through them. It is thought that large-scale systems can be constructed of adaptive building blocks. These adaptive systems would be flexible in function, reliable and would resist severe damage characteristics of living creatures. Neuron models were tested by interconnecting them into various networks to perform simple control tasks. The test results were evaluated and the evaluation used to improve the theory and the neuron model. Two basic analysis methods were used to study neuromime networks: a sequential machine analysis and an optimal process method applying Pontryagin's maximum principle. The sequential analysis method proved unsatisfactory when applied to an attempted description of an adjustment rule for the neuron model. This difficulty led to an application of optimal processes. The application of Pontryagin's maximum principle to the analysis of the neuron model network described both optimum conditions for a system and criteria useful for developing the adjustment rule.
Provenance: RAF Centre of Aviation Medicine