Energy-Aware ML for Embedded Systems
The mind map focuses on "Energy-Aware Machine Learning for Embedded Systems" and is divided into several key branches. Optimization Techniques include methods like pruning, which reduces network size by eliminating unnecessary neurons, quantization, which lowers precision to save energy and memory, hardware acceleration using specialized processors like TPUs and GPUs, and neuromorphic computing that mimics brain functions for high efficiency. Hardware Platforms explored are microcontrollers (...
Energy-Aware ML for Embedded Systems
da Sahil Guglani
