Abstract: The evolution of deep neural networks (DNNs) naturally leads to an increase in model size. This necessitates various model compression techniques, such as pruning and quantization, to reduce ...