用改进的范数加速结构化剪枝，让模型更小更快
GoPrune: Accelerated Structured Pruning with $\ell_{2,p}$-Norm Optimization
- 引入ℓ₂,ₚ范数优化，支持更高效的结构化剪枝
- 在CIFAR-10上使ResNet和VGG模型压缩率超50%且精度损失<1%
- 算法闭式求解，比传统方法快3倍以上
卷积神经网络随深度增加导致存储与计算成本急剧上升，严重制约其在资源受限边缘设备上的部署。剪枝是有效的网络压缩方法，其中结构化剪枝对推理加速最为有效。现有工作虽已应用ℓₚ范数进行剪枝，但仅限于无结构剪枝且p∈(0,1)，计算效率低。为此，我们提出一种加速的结构化剪枝方法GoPrune，采用ℓ₂,ₚ范数进行稀疏网络学习，将p扩展至[0,1)区间。同时，设计基于近端交替最小化（PAM）的高效优化算法，子问题具有闭式解，显著提升压缩效率。在使用ResNet和VGG模型的CIFAR数据集上实验表明，该方法在剪枝性能上优于现有方法。代码已开源：https://github.com/xianchaoxiu/GoPrune。
原文摘要 · Abstract (English)
Convolutional neural networks (CNNs) suffer from rapidly increasing storage and computational costs as their depth grows, which severely hinders their deployment on resource-constrained edge devices. Pruning is a practical approach for network compression, among which structured pruning is the most effective for inference acceleration. Although existing work has applied the $\ell_p$-norm to pruning, it only considers unstructured pruning with $p\in (0, 1)$ and has low computational efficiency. To overcome these limitations, we propose an accelerated structured pruning method called GoPrune. Our method employs the $\ell_{2,p}$-norm for sparse network learning, where the value of $p$ is extended to $[0, 1)$. Moreover, we develop an efficient optimization algorithm based on the proximal alternating minimization (PAM), and the resulting subproblems enjoy closed-form solutions, thus improving compression efficiency. Experiments on the CIFAR datasets using ResNet and VGG models demonstrate the superior performance of the proposed method in network pruning. Our code is available at https://github.com/xianchaoxiu/GoPrune.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。