Home // BRAININFO 2024, The Ninth International Conference on Neuroscience and Cognitive Brain Information // View article

A Self-Learning Neuromorphic System

Authors:
Rory Lewis
Michael Bihn
Daniel Barbotko
Zhenqi Liu

Keywords: Neuromorphic, AI, Cmputational Neuroscience.

Abstract:
In the continuing research to implement a plurality of self-wiring synapses comprised of Field Programmable Gate Arrays (FPGAs) on a Complementary Metal Oxide Semiconductor (CMOS) system to accommodate artificial intelligence (AI) on a microprocessor, we delve into how a system can emulate not just a self-wiring CMOS system, but also how it can emulate brain growth at the connectome level. The enigma of contemporary advancements in AI and chip manufacturing diverging from bio-inspired systems is fascinating, especially given that AI and microprocessor engineers readily acknowledge the superior capabilities of biological brains. This paper introduces a bio-inspired device made of steel, plastic, and silica, which autonomously rewires itself, evolving and enhancing its intelligence without human intervention. The research will delve into the intricacies of the FPGA prototype's functionality, shedding light on both its technical aspects and the broader social and technological implications associated with the development of this neuromorphic chip. Next, we introduce the theoretical ability for the CMOS to grow its connectivity to FPGAs as does a human baby. Herein we introduce the uniqueness of applying the logistical growth function to the curve fitting of the multidimensional measures of brain growth, on a CMOS system.

Pages: 11 to 16

Copyright: Copyright (c) IARIA, 2024

Publication date: March 10, 2024

Published in: conference

ISSN: 2519-8653

ISBN: 978-1-68558-127-5

Location: Athens, Greece

Dates: from March 10, 2024 to March 14, 2024