Self-organized Cognitive Algebraic Neural Network

Prabir Sen

2020

Abstract

This paper refers to author’s patented invention that introduces a more efficient statistical (machine) learning method. Inspired by neuroscience, the paper combines the synaptic networks and graphs of quantum network to constitute interactions as information flow. Hitherto, several machine learning algorithms had some influence in business decision-making under uncertainty, however the dynamic cognitive states and differences thereof, at different timepoints, play an important role in transactional businesses to derive choice and choice-sets for decision-making at societal scale. In addition, deep neural functions that reflect the direction of information flow, the cliques and cavities, necessitate a new computational framework and deeper learning method. This paper introduces a proactive-retroactive learning technique - a quantified measurement of a multi-layered-multi-dimensional architecture based on a Self-Organized Cognitive Algebraic Neural Network (SCANN) integrated with Voronoi geometry – to deduce the optimal (cognitive) state, action, response and reward (pay-off) in more realistic imperfect and incomplete information conditions. This quantified measurement of SCANN produced an efficient and optimal learning results for individuals’ transactional activities and for nearest-neighbor, as a group, for which the individual is a member. This paper also discusses and characterizes SCANN for those who handle decisions under conditions of uncertainty, juxtaposed between human and machine intelligence.

Paper Citation

in Harvard Style

Sen P. (2020). Self-organized Cognitive Algebraic Neural Network. In Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-395-7, pages 836-845. DOI: 10.5220/0009141408360845

in Bibtex Style

@conference{icaart20,
author={Prabir Sen},
title={Self-organized Cognitive Algebraic Neural Network},
booktitle={Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2020},
pages={836-845},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0009141408360845},
isbn={978-989-758-395-7},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Self-organized Cognitive Algebraic Neural Network
SN - 978-989-758-395-7
AU - Sen P.
PY - 2020
SP - 836
EP - 845
DO - 10.5220/0009141408360845