Optimization of Spiking Neural Networks for Radar Applications

Více o knize

Edge devices play a crucial role in enhancing Internet of Things (IoT) applications by enabling real-time processing, reducing latency, and improving security and scalability. As AI adoption at the edge rises, the need for efficient data processing drives innovation. The book critically examines traditional deep neural networks, particularly their energy efficiency during inference, and explores methods to reduce energy consumption, such as tiny networks and weight quantization. It also addresses the challenges of AI accelerators' physical size, focusing on spiking neural networks (SNNs) for radar data processing and optimized algorithms.