Jawaharlal Nehru Technological University Researcher Updates Current Study Findings on Emerging Technologies (Efficient Hardware Implementation of Spiking Neural Networks Using Nonstandard Finite Difference Scheme for Leaky Integrate and Fire...).

A recent research paper from Jawaharlal Nehru Technological University in India focuses on implementing efficient Spiking Neural Networks (SNNs) using Field-Programmable Gate Array (FPGA) technology. The researchers optimized the mathematical equations of the Leaky Integrate and Fire (LIF) neuron model by approximating nonlinear functions, resulting in a simple, cost-effective, and high-speed design. The proposed model achieved a classification accuracy of 97.8% for basic pattern recognition using large datasets, and an 84% accuracy rate for COVID-19 chest CT scan image classification, outperforming existing SNNs. The study highlights the importance of discrete versions of continuous models for hardware implementation in emerging technologies. [Extracted from the article]