What is it about?
Mental phenomena based models are on a different scale from pulsing models. Availability of the pulsing model makes it possible to search for mechanistic explanations of the mental phenomenon. To maintain objective validity for any inferred mechanistic explanations the pulsing neural network must maintain continuity of knowledge with the original model. The paper develops an adaptive pulse-coded neural network using model-reference principle. The methods used in the development process helps maintain continuity of knowledge among the different model scales.
Featured Image
Photo by Alina Grubnyak on Unsplash
Why is it important?
Although model-reference principle is tool known in control systems it not known to be implemented in developing models in neuroscience. We therefore demonstrate an approach novel to modelling in computational neuroscience. In addition, set-membership theory is used as a basis for comparing the different signal types. This is also a technique known mostly within the information theorist community. We therefore demonstrate the application of tools that assist in modelling, which is not developed ad-hoc but maintains continuity of knowledge. The demonstration is done from larger-scale to smaller-scale model, i.e., scientific reduction. However, the tools can also be used for model-order reduction.
Perspectives
This paper shows that model-reference principle can be a powerful tool to implement model-order reduction and scientific reduction in computational neuroscience.
Dr Lungsi Sharma
Ronin Institute
Read the Original
This page is a summary of: A demonstration of using the model reference principle to develop the function-oriented adaptive pulse-coded neural network, SIMULATION, July 2019, SAGE Publications,
DOI: 10.1177/0037549719860587.
You can read the full text:
Resources
Contributors
The following have contributed to this page
