On the study and development of biological plausible computational intelligent models

• Grant number: 23/01374-3
• Support Opportunities: Regular Research Grants
• Start date: July 01, 2023
• End date: June 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Agreement: CNPq
• Principal Investigator: João Paulo Papa
• Grantee: João Paulo Papa
• Host Institution: Faculdade de Ciências (FC). Universidade Estadual Paulista (UNESP). Campus de Bauru. Bauru , SP, Brazil
• Associated research grant: 13/07375-0 - CeMEAI - Center for Mathematical Sciences Applied to Industry, AP.CEPID
• Associated scholarship(s): 23/10823-6 - On the Study and Development of Biological Plausible Computational Intelligent Models, BP.JD

Abstract

Recent neurological discoveries on the cortex, the hippocampus, and different areas of the brain shed light on pyramidal cells, a neuronal architecture composed of five layers, i.e., Soma, Basal dendrites, Apical dendrites, Axon, and Collateral axon, whose interaction is responsible for the forward and backward flow of information, as well as the integration among context and memory, among other tasks. Such discoveries inspired the development of more biologically plausible intelligent computational models, usually implying more accurate and efficient algorithms. Regarding such models, one can cite contextually guided approaches, which mimic pyramidal cells' behavior by using contextual information to deal with ambiguity, implementing mechanisms to tackle temporal information and simulate the memory. Other studies go beyond dealing with the credit assignment problem, i.e., assign each connection in a neural network a proper adjustment based on its influence on the output through the primary principles of pyramidal neurons. In this context, burst-dependent learning or Burstpropagation provides a paradigm based on such principles, which can be implemented on fully-connected, convolutional, and spiking neural networks, reaching state-of-the-art results. This project aims to implement existing models and develop more biologically plausible machine learning architectures, as well as apply such techniques to tackle problems in diverse research fields, like medicine and engineering. (AU)