Parallel computing

Q-EEGNET: an Energy-Efficient 8-bit Quantized Parallel EEGNet Implementation for Edge Motor-Imagery Brain-Machine Interfaces
Q-EEGNET is an energy-efficient implementation of EEGNET, a compact Convolutional Neural Network suitable for many BMI paradigms. Our algorithmic and implementation optimizations improve energy efficiency by more than 250 times.
Tibor Schneider, Xiaying Wang, Michael Hersche, Lukas Cavigelli, Luca Benini
PDF Cite Code DOI