Separate, Dynamic and Differentiable (SMART) Pruner for Block/Output Channel Pruning on Computer Vision Tasks

arXiv:2403.19969v1 Announce Type: cross
Abstract: Deep Neural Network (DNN) pruning has emerged as a key strategy to reduce model size, improve inference latency, and lower power consumption on DNN accelerators. Among various pruning techniques, block and output channel pruning have shown significant potential in accelerating hardware performance. However, their accuracy often requires further improvement. In response to this challenge, we introduce a separate, dynamic and differentiable (SMART) pruner. This pruner stands out by utilizing a separate, learnable probability mask for …

abstract accelerators arxiv block computer computer vision consumption cs.cv cs.lg deep neural network differentiable dnn dnn accelerators dynamic hardware inference inference latency key latency network neural network power power consumption pruning reduce smart strategy tasks type vision