Salvato in:
Dettagli Bibliografici
Autori principali: Shang, Ronghua, Zhu, Songling, Wu, Yinan, Zhang, Weitong, Jiao, Licheng, Xu, Songhua
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:https://arxiv.org/abs/2303.16212
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916077287505920
author Shang, Ronghua
Zhu, Songling
Wu, Yinan
Zhang, Weitong
Jiao, Licheng
Xu, Songhua
author_facet Shang, Ronghua
Zhu, Songling
Wu, Yinan
Zhang, Weitong
Jiao, Licheng
Xu, Songhua
contents Model compression plays a vital role in the practical deployment of deep neural networks (DNNs), and evolutionary multi-objective (EMO) pruning is an essential tool in balancing the compression rate and performance of the DNNs. However, due to its population-based nature, EMO pruning suffers from the complex optimization space and the resource-intensive structure verification process, especially in complex networks. To this end, a multi-objective complex network pruning framework based on divide-and-conquer and global performance impairment ranking (EMO-DIR) is proposed in this paper. Firstly, a divide-and-conquer EMO network pruning method is proposed, which decomposes the complex task of EMO pruning on the entire network into easier sub-tasks on multiple sub-networks. On the one hand, this decomposition narrows the pruning optimization space and decreases the optimization difficulty; on the other hand, the smaller network structure converges faster, so the proposed algorithm consumes lower computational resources. Secondly, a sub-network training method based on cross-network constraints is designed, which could bridge independent EMO pruning sub-tasks, allowing them to collaborate better and improving the overall performance of the pruned network. Finally, a multiple sub-networks joint pruning method based on EMO is proposed. This method combines the Pareto Fronts from EMO pruning results on multiple sub-networks through global performance impairment ranking to design a joint pruning scheme. The rich experiments on CIFAR-10/100 and ImageNet-100/1k are conducted. The proposed algorithm achieves a comparable performance with the state-of-the-art pruning methods.
format Preprint
id arxiv_https___arxiv_org_abs_2303_16212
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle A Multi-objective Complex Network Pruning Framework Based on Divide-and-conquer and Global Performance Impairment Ranking
Shang, Ronghua
Zhu, Songling
Wu, Yinan
Zhang, Weitong
Jiao, Licheng
Xu, Songhua
Machine Learning
Artificial Intelligence
Model compression plays a vital role in the practical deployment of deep neural networks (DNNs), and evolutionary multi-objective (EMO) pruning is an essential tool in balancing the compression rate and performance of the DNNs. However, due to its population-based nature, EMO pruning suffers from the complex optimization space and the resource-intensive structure verification process, especially in complex networks. To this end, a multi-objective complex network pruning framework based on divide-and-conquer and global performance impairment ranking (EMO-DIR) is proposed in this paper. Firstly, a divide-and-conquer EMO network pruning method is proposed, which decomposes the complex task of EMO pruning on the entire network into easier sub-tasks on multiple sub-networks. On the one hand, this decomposition narrows the pruning optimization space and decreases the optimization difficulty; on the other hand, the smaller network structure converges faster, so the proposed algorithm consumes lower computational resources. Secondly, a sub-network training method based on cross-network constraints is designed, which could bridge independent EMO pruning sub-tasks, allowing them to collaborate better and improving the overall performance of the pruned network. Finally, a multiple sub-networks joint pruning method based on EMO is proposed. This method combines the Pareto Fronts from EMO pruning results on multiple sub-networks through global performance impairment ranking to design a joint pruning scheme. The rich experiments on CIFAR-10/100 and ImageNet-100/1k are conducted. The proposed algorithm achieves a comparable performance with the state-of-the-art pruning methods.
title A Multi-objective Complex Network Pruning Framework Based on Divide-and-conquer and Global Performance Impairment Ranking
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2303.16212