用分数阶高斯滤波与剪枝压缩模型，轻量高效且精度损失小。
FGFP: A Fractional Gaussian Filter and Pruning for Deep Neural Networks Compression
- 引入分数阶微分与高斯函数构建轻量滤波器，参数仅7个
- 在CIFAR-10上压缩85.2%仍保持98.48%准确率，ImageNet上压缩69.1%
- 适合部署在资源受限的边缘设备，兼顾精度与效率
网络压缩技术近年来愈发重要，因深度神经网络（DNN）对真实场景中边缘设备的负载过重。尽管已有多种方法压缩网络参数，但部署仍具挑战。为此，我们提出分数阶高斯滤波与剪枝（FGFP）框架，融合分数阶微分与高斯函数构造分数阶高斯滤波器（FGFs）。为降低分数阶微分运算复杂度，引入Grünwald-Letnikov分数阶导数近似微分方程。每个核的参数量降至仅7个。除架构优化外，还结合自适应非结构化剪枝（AUP），实现更高压缩率。多架构与基准测试表明，本框架在精度与压缩比上均优于近期方法：在CIFAR-10上，ResNet-20仅损失1.52%准确率，模型大小减少85.2%；在ImageNet2012上，ResNet-50仅损失1.63%准确率，模型大小减少69.1%。
原文摘要 · Abstract (English)
Network compression techniques have become increasingly important in recent years because the loads of Deep Neural Networks (DNNs) are heavy for edge devices in real-world applications. While many methods compress neural network parameters, deploying these models on edge devices remains challenging. To address this, we propose the fractional Gaussian filter and pruning (FGFP) framework, which integrates fractional-order differential calculus and Gaussian function to construct fractional Gaussian filters (FGFs). To reduce the computational complexity of fractional-order differential operations, we introduce Grünwald-Letnikov fractional derivatives to approximate the fractional-order differential equation. The number of parameters for each kernel in FGF is minimized to only seven. Beyond the architecture of Fractional Gaussian Filters, our FGFP framework also incorporates Adaptive Unstructured Pruning (AUP) to achieve higher compression ratios. Experiments on various architectures and benchmarks show that our FGFP framework outperforms recent methods in accuracy and compression. On CIFAR-10, ResNet-20 achieves only a 1.52% drop in accuracy while reducing the model size by 85.2%. On ImageNet2012, ResNet-50 achieves only a 1.63% drop in accuracy while reducing the model size by 69.1%.
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。