FGFP: A Fractional Gaussian Filter and Pruning for Deep Neural Networks Compression
This work addresses the challenge of deploying heavy DNNs on edge devices, offering a novel compression method with strong specific gains.
The paper tackles the problem of compressing deep neural networks for deployment on edge devices by proposing the FGFP framework, which integrates fractional Gaussian filters and adaptive unstructured pruning, achieving a 1.52% accuracy drop with 85.2% size reduction on CIFAR-10 and a 1.63% drop with 69.1% reduction on ImageNet.
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%.