Future of Neural Networks and Energy Consumption Aspects

In this position paper, we provide a comprehensive presentation of the current state of Neural Networks (NNs), focusing specifically on their two primary branches: Artificial Neural Networks (ANNs) and Spiking Neural Networks (SNNs). We commence with a historical review of the development of these two significant directions within the connections paradigm. This evaluation will include a detailed exploration of the prevailing problems impacting current ANNs and SNNs methodologies. Subsequently, we share our perspective on the potential resolution of some of these challenges. Our approach leverages biological SNN analogy methods and implements bio-SNN phenomena in leading-edge Artificial Intelligence technologies.