Analogical Study of Activation Concept in Neural Networks with Neat- Python Module

• Published in: Revue d intelligence artificielle Vol. 37; no. 2; p. 249
• Main Authors: Kaul, Nishit, Kaul, Sameer, Zaman, Majid, Bakshi, Waseem Jeelani, Sheikh Amir Fayaz
• Format: Journal Article
• Language: English
• Published: Edmonton International Information and Engineering Technology Association (IIETA) 01.04.2023
• Subjects: Artificial intelligence; Complexity; Computer & video games; Computer vision; Engines; Genetic algorithms; Machine learning; Modules; Mutation; Natural language processing; Neural networks; Real time; Robotics; Simulators; Workflow
• ISSN: 0992-499X; 1958-5748; 1958-5748
• DOI: 10.18280/ria.370201

Determining real-time machine simulation and functionalities of complex AI Engines is difficult to comprehend and is rarely discussed. We present a technique to analyze the workflow of one such engine, the NEAT engine, one of the fundamental and robust training engines in the current machine learning scenario. Computer Vision also presents a great approach towards working in real-time speedy functioning virtual simulators and visual platforms, whereas NEAT is not specialized in the same, but can perform to the best of the extent in the present day Machine Learning aspect. Technologies like Python, PyGame, and CsV were used to facilitate the research. So far, we have tested both the frameworks on real time cases, and it is safe to say that the NEAT module has presented an accurate trajectory, besides greater time complexity. Thus, we not only evaluate the accuracy but the other key factors as well. This study demonstrates that NEAT has the ability to address other difficult issues in the future and can produce excellent outcomes with a relatively small population. Robotics, artificial intelligence for video games, natural language processing, and healthcare are some of the potential future applications for NEAT.