paper-with-me

홈 › Papers

Group Fisher Pruning for Practical Network Compression

2021-08-02 · Liyang Liu, Shilong Zhang, Zhanghui Kuang, Aojun Zhou, Jing-Hao Xue, Xinjiang Wang, Yimin Chen, Wenming Yang, Qingmin Liao, Wayne Zhang

Network compression has been widely studied since it is able to reduce the memory and computation cost during inference. However, previous methods seldom deal with complicated structures like residual connections, group/depth-wise convolution and feature pyramid network, where channels of multiple layers are coupled and need to be pruned simultaneously. In this paper, we present a general channel pruning approach that can be applied to various complicated structures. Particularly, we propose a layer grouping algorithm to find coupled channels automatically. Then we derive a unified metric based on Fisher information to evaluate the importance of a single channel and coupled channels. Moreover, we find that inference speedup on GPUs is more correlated with the reduction of memory rather than FLOPs, and thus we employ the memory reduction of each channel to normalize the importance. Our method can be used to prune any structures including those with coupled channels. We conduct extensive experiments on various backbones, including the classic ResNet and ResNeXt, mobile-friendly MobileNetV2, and the NAS-based RegNet, both on image classification and object detection which is under-explored. Experimental results validate that our method can effectively prune sophisticated networks, boosting inference speed without sacrificing accuracy.

📄 PDF Abstract BibTeX arXiv:2108.00708

Code (2)

jshilong/FisherPruning 공식 구현 pytorch
Ben-Louis/FisherPruning-Pytorch pytorch

Tasks

image-classificationImage ClassificationNetwork Pruningobject-detectionObject Detection

Methods 이 논문이 사용한 방법론

Pruning 설명 없음
Pointwise Convolution Pointwise Convolution is a type of convolution that uses a 1x1 kernel: a kernel that iterates through every single point. This…
Depthwise Convolution Depthwise Convolution is a type of convolution where we apply a single convolutional filter for each input channel. In the regular 2D…
Residual Connection 설명 없음
ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…
Average Pooling 설명 없음
Grouped Convolution A Grouped Convolution uses a group of convolutions - multiple kernels per layer - resulting in multiple channel outputs per layer. This leads to wider networks helping a…
Depthwise Separable Convolution While standard convolution performs the channelwise and spatial-wise computation in one step, Depthwise Separable Convolution …

Similar Papers 제목 키워드 기반

WoodFisher: Efficient Second-Order Approximation for Neural Network Compression

2020-04-29 · NeurIPS 2020 12 · Sidak Pal Singh, Dan Alistarh

Second-order information, in the form of Hessian- or Inverse-Hessian-vector products, is a fundamental tool for solving optimization problems. Recently, there has been significant interest in utilizing this information i…

image-classificationImage ClassificationModel CompressionNeural Network Compression

F-WANDA: Fisher-Reweighted Post-Training Pruning for Sustainable Deployment of Large Language Models

2026-08-01 · Himanshu Mishra arxiv

One-shot post-training pruning is the most energy-frugal compression strategy for largelanguage models (LLMs), yet existing approaches trade either quality (WANDA) or compute cost (SPARSEGPT). We introduce F-WANDA, a dro…

Component-Aware Pruning Framework for Neural Network Controllers via Gradient-Based Importance Estimation

2026-01-27 · Ganesh Sundaram, Jonas Ulmen, Daniel Görges arxiv

The transition from monolithic to multi-component neural architectures in advanced neural network controllers poses substantial challenges due to the high computational complexity of the latter. Conventional model compre…

Model Compression

Hierarchical Reinforcement Learning for Neural Network Compression (HiReLC): Pruning and Quantization

2026-06-24 · Kamar Hibatallah Baghdadi, Kawther Guoual Belhamidi, Sara Belhadj, Aissa Boulmerka 외 arxiv

We present HiReLC, a hierarchical ensemble-reinforcement learning framework for automated joint quantization and structured pruning of deep neural networks. The framework decomposes the compression search across two leve…

Hierarchical Reinforcement LearningNeural Network CompressionActive Learning

Towards Optimal Compression: Joint Pruning and Quantization

2023-02-15 · Ben Zandonati, Glenn Bucagu, Adrian Alan Pol, Maurizio Pierini 외

Model compression is instrumental in optimizing deep neural network inference on resource-constrained hardware. The prevailing methods for network compression, namely quantization and pruning, have been shown to enhance …

Model CompressionNeural Architecture SearchQuantization