计算机应用与软件2024，Vol.41Issue(1)：253-260,8.DOI:10.3969/j.issn.1000-386x.2024.01.037
基于聚类和生成对抗学习模型的滤波器剪枝
FPCC-GAN:CLUSTER CENTER AND GENERATIVE ADVERSARIAL LEARNING IN FILTER LEVEL PRUNING
摘要
Abstract
The deep architecture and parameter redundancy of deep neural network will lead to high computational cost.Deep neural network compression and acceleration has become an important issue in recent years.To address the norm-criterion limitation and label dependence of current methods,we propose a structured filter pruning method based on cluster center and generative adversarial learning(FPCC-GAN).(1)The filters were clustered by K-means clustering algorithm for every convolution layer.(2)Filters closer to the cluster center were pruned proportionally,which extracted redundant features.(3)Generative adversarial learning was used for iteratively training.The experimental results show that compared with current mainstream methods,the proposed method has higher accuracy.关键词
网络压缩/深度神经网络加速/参数剪枝/聚类/生成对抗学习Key words
Network compression/Network acceleration/Parameter pruning/Clustering/Generative adversarial learning分类
信息技术与安全科学引用本文复制引用
冯叶棋,张俊三,邵明文,张世栋..基于聚类和生成对抗学习模型的滤波器剪枝[J].计算机应用与软件,2024,41(1):253-260,8.基金项目
国家自然科学基金项目(61673396) （61673396）
中央高校基本科研业务费专项资金项目(20CX05019A) （20CX05019A）
中石油重大科技项目(ZD2019-183-004). （ZD2019-183-004）
