Exploring Quantization and Mapping Synergy in Hardware-Aware Deep Neural Network Accelerators

2 code implementations • 8 Apr 2024 • Jan Klhufek, Miroslav Safar, Vojtech Mrazek, Zdenek Vasicek, Lukas Sekanina

Energy efficiency and memory footprint of a convolutional neural network (CNN) implemented on a CNN inference accelerator depend on many factors, including a weight quantization strategy (i. e., data types and bit-widths) and mapping (i. e., placement and scheduling of DNN elementary operations on hardware units of the accelerator).

Quantization Scheduling