用统计方法筛选关键参数，让视觉神经网络变小10倍仍保持精度。
Exploring Vision Neural Network Pruning via Screening Methodology
- 基于F统计量和加权评估，量化连接与通道重要性
- 在多个真实数据集上实现存储与计算降低一个数量级
- 适用于各类卷积和全连接网络，兼顾结构化与非结构化剪枝
现代深度神经网络（DNN）凭借数十亿参数实现了卓越性能，但其巨大规模带来了高昂的存储与计算开销，限制了在边缘设备上的实时、节能部署。本文提出一种网络剪枝框架，通过分类类别层面的组件重要性统计分析，剔除非必要参数，在不损失模型精度的前提下，使存储与计算需求降低一个数量级。方法采用基于F统计量的筛选技术结合加权评估机制，量化连接与通道贡献，统一支持非结构化与结构化剪枝。在涵盖全连接网络（FNNs）与卷积神经网络（CNNs）的真实视觉数据集上的大量实验表明，该框架可生成紧凑高效的模型，性能媲美当前最优方法。
原文摘要 · Abstract (English)
The remarkable performance of modern deep neural networks (DNNs) is largely driven by their massive scale, often comprising tens to hundreds of millions-or even billions-of parameters. However, such a scale incurs substantial storage and computational costs, hindering deployment on platforms such as edge devices that require energy-efficient and real-time processing. In this paper, we propose a network pruning framework that reduces both storage and computation requirements by an order of magnitude while preserving model accuracy. Our approach eliminates non-essential parameters through a statistical analysis of component significance across classification categories. Specifically, we employ a F-statistic-based screening technique combined with a weighted evaluation scheme to quantify the contributions of connections and channels, enabling both unstructured and structured pruning within a unified framework. Extensive experiments on real-world vision datasets, covering both fully connected neural networks (FNNs) and convolutional neural networks (CNNs), demonstrate that the proposed framework produces compact and efficient models that are highly competitive with the state of art apporoaches.
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。