Hardware-Aware DNN Compression via Diverse Pruning and Mixed-Precision Quantization

no code implementations • 23 Dec 2023 • Konstantinos Balaskas, Andreas Karatzas, Christos Sad, Kostas Siozios, Iraklis Anagnostopoulos, Georgios Zervakis, Jörg Henkel

We explore, for the first time, per-layer fine- and coarse-grained pruning, in the same DNN architecture, in addition to low bit-width mixed-precision quantization for weights and activations.

Quantization Reinforcement Learning (RL)