.png){kind=icon}
返回
Coresets based asynchronous network slimming
.png){kind=icon}
OA
AI
DOI：10.1007/s10489-022-04092-0.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Pruning is effective to reduce neural networks' parameters and accelerate inferences, facilitating deep learning in resource-limited scenarios. This paper proposes an asynchronous pruning method for multi-branch networks on the basis of our previous work on channel coresets constructions, to achieve module-level pruning. Firstly, this paper accelerates coreset based pruning by batch sampling with a sampling probability decided on our-designed importance function. Secondly, this paper gives asynchronous pruning solutions with an in-place distillation of feature maps for deployment on multi-branch networks such as ResNet and SqueezeNet. Thirdly, this paper provides an extension to neuron pruning by grouping weights as channels. During tests on sensitivity of different layers to channel pruning, our method outperforms comparison schemes on object detection networks, indicating advantages of data-independent channel selections in maintaining precision. As shown in tests of asynchronous pruning solutions on multi-branch classification networks, our method further decreases FLOPs with a small accuracy decline on ResNet and acquires a small accuracy increment on SqueezeNet. In tests on neuron pruning, our method achieves an accuracy comparable to existing coreset based pruning methods by two solutions of precision recovery.
Keyword:
Channel pruning
Coreset theory
Knowledge distillation
Data-independent pruning
期刊
IF:
3.5
论文数:
7.6K
被引数:
1.7W
机构
引用论文
Calcium-modulating cyclophilin ligand regulates membrane trafficking of postsynaptic GABAA receptors
OAAI
OAAI分享收藏Pruning filters with L1-norm and capped L1-norm for CNN compression具有L1-norm和上限L1-norm的修剪过滤器，用于CNN压缩.png)
APPLIED INTELLIGENCE
APPLIED INTELLIGENCEIF3.5
PREAI
分享收藏Colorimetric detection of pesticides based on calixarene modified silver nanoparticles in water
PREAI
分享收藏
查看更多内容