Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning
arXiv:2607.11473 · doi:10.1109/DAC56929.2023.10247965
The paper introduces TECO, a framework that jointly prunes depth, width, and input resolution of convolutional neural networks to improve speed and resource usage on embedded devices while maintaining accuracy.
Abstract
In this paper, we propose TECO, a multi-dimensional pruning framework to collaboratively prune the three dimensions (depth, width, and resolution) of convolutional neural networks (CNNs) for better execution efficiency on embedded hardware. In TECO, we first introduce a two-stage importance evaluation framework, which efficiently and comprehensively evaluates each pruning unit according to both the local importance inside each dimension and the global importance across different dimensions. Based on the evaluation framework, we present a heuristic pruning algorithm to progressively prune the three dimensions of CNNs towards the optimal trade-off between accuracy and efficiency. Experiments on multiple benchmarks validate the advantages of TECO over existing state-of-the-art (SOTA) approaches. The code and pre-trained models are available at https://github.com/ntuliuteam/Teco.
Author's accepted version. Published in Proceedings of the 60th ACM/IEEE Design Automation Conference (DAC 2023)