融合剪枝与量化，用专家路由提升压缩后模型性能。
Hybrid Compression: Integrating Pruning and Quantization for Optimized Neural Networks
- 先剪枝量化降规模，再用专家混合路由优化推理。
- 压缩后模型参数和计算量大幅下降，精度损失极小。
- 适合资源受限设备部署，尤其对边缘计算友好。
深度神经网络近年来取得显著进展，并广泛应用于各类场景。然而，在嵌入式和边缘设备上训练与部署模型时，受限于内存和算力，面临严峻挑战。神经网络压缩可在模型大小与性能间实现权衡。本文提出一种两阶段模型压缩新方法：首先使用剪枝和量化技术显著缩小模型规模；随后引入专家混合（Mixture of Experts, MoE）机制，将压缩后的多个小型专家模型进行路由，以在保持推理效率的同时提升性能。MoE由多个中等规模的专家模型组成，可稳定输出。在多个基准数据集上的实验表明，该方法成功压缩了卷积神经网络（CNN）模型，在参数和浮点运算量（FLOPs）大幅降低的同时，精度下降可忽略不计。
原文摘要 · Abstract (English)
Deep neural networks have witnessed remarkable advancements in recent years and have become integral to various applications. However, alongside these developments, training and deployment of neural network models on embedding and edge devices face significant challenges due to limited memory and computational resources. These problems can be addressed with deep neural network compression, which involves a trade-off between model size and performance. In this paper, we propose a novel method for model compression through two phases. First, we utilize model compression techniques, such as pruning and quantization, to significantly reduce the model size. Then, we use Mixture of Experts to route the previously compressed models to enhance performance while maintaining a balance in inference efficiency. MoEs consist of multiple expert models (i.e., compressed models) that are moderately sized and deliver stable performance. Experimental results on several benchmark datasets show that our method successfully compresses CNN models which achieves substantial reductions in FLOPs and parameters with a negligible accuracy drop.
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。