[Paper Review] Neural Networks, Artificial Intelligence and the Computational Brain

Martin C. Nwadiugwu|arXiv (Cornell University)|Dec 25, 2020

EEG and Brain-Computer Interfaces14 references24 citations

TL;DR

This paper explores artificial neural networks (ANNs) as computational models inspired by biological neural networks, emphasizing their role in advancing artificial intelligence through brain-inspired architectures. It compares ANNs to traditional computing frameworks, highlighting their applications in anomaly detection, medical diagnostics, and pattern recognition, while advocating for brain-like intelligence as a foundation for next-generation AI systems.

ABSTRACT

In recent years, several studies have provided insight on the functioning of the brain which consists of neurons and form networks via interconnection among them by synapses. Neural networks are formed by interconnected systems of neurons, and are of two types, namely, the Artificial Neural Network (ANNs) and Biological Neural Network (interconnected nerve cells). The ANNs are computationally influenced by human neurons and are used in modelling neural systems. The reasoning foundations of ANNs have been useful in anomaly detection, in areas of medicine such as instant physician, electronic noses, pattern recognition, and modelling biological systems. Advancing research in artificial intelligence using the architecture of the human brain seeks to model systems by studying the brain rather than looking to technology for brain models. This study explores the concept of ANNs as a simulator of the biological neuron, and its area of applications. It also explores why brain-like intelligence is needed and how it differs from computational framework by comparing neural networks to contemporary computers and their modern day implementation.

Motivation & Objective

• To examine the conceptual and functional parallels between artificial neural networks (ANNs) and biological neural networks in the brain.
• To investigate why brain-inspired intelligence is essential for advancing artificial intelligence beyond traditional computational models.
• To compare the architectural and operational differences between ANNs and conventional computers in processing information.
• To evaluate the practical applications of ANNs in real-world domains such as healthcare, pattern recognition, and electronic sensing.
• To establish a foundation for future AI research by modeling systems based on neural system principles rather than technological constraints.

Proposed method

• The paper uses a conceptual and comparative analysis to model ANNs as simulations of biological neurons, focusing on their interconnection via synapses.
• It draws on recent neuroscientific findings to inform the design and functionality of ANNs, emphasizing biological plausibility.
• The study compares the information-processing mechanisms of ANNs with those of traditional digital computers, highlighting differences in parallelism and adaptability.
• It reviews existing applications of ANNs in medicine (e.g., instant physician, electronic noses), anomaly detection, and biological system modeling.
• The framework is built on the premise that intelligence emerges from networked, adaptive, and interconnected processing units—mirroring neural systems.
• The paper employs a qualitative synthesis of neuroscience and AI literature to support its argument for brain-inspired AI development.

Experimental results

Research questions

• RQ1 How do artificial neural networks emulate the structure and function of biological neural networks in the human brain?
• RQ2 What are the key differences between artificial neural networks and conventional computing architectures in terms of information processing?
• RQ3 Why is brain-like intelligence considered essential for advancing artificial intelligence beyond current technological paradigms?
• RQ4 In what ways do ANNs enable improved performance in anomaly detection, medical diagnostics, and pattern recognition?
• RQ5 How can modeling AI systems after neural networks lead to more adaptive and biologically plausible intelligence?

Key findings

• Artificial neural networks are computationally inspired by the human brain’s interconnected neurons and synapses, enabling them to model complex neural systems.
• ANNs demonstrate strong performance in anomaly detection, particularly in medical applications such as instant physician systems and electronic noses.
• The paper establishes that brain-inspired architectures offer advantages in adaptability and parallel processing over traditional digital computing models.
• The integration of ANNs into biological system modeling enables more accurate simulations of neural behavior and cognitive functions.
• The study concludes that future AI development should prioritize neural network architectures over technology-driven models to achieve more robust and generalizable intelligence.
• The conceptual framework supports the use of ANNs as a viable simulator of biological neurons, with broad applicability across neuroscience and AI research.