PAC-Bayesian framework based drop-path method for 2D discriminative convolutional network pruning

DOI10.1007/S11045-019-00686-ZMaRDI QIDQ2192902zbMATH OpenOpenAlexWikidataFDO

Authors Qinghe Zheng, Xinyu Tian, Mingqiang Yang, Yulin Wu, Huake Su

Publication date 21 August 2020

Published in Multidimensional Systems and Signal Processing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/s11045-019-00686-z

zbMATH Keywords

deep learningmodel compressionstructure explorationmodel accelerationgeneralization error boundary

Mathematics Subject Classification ID

Bayesian inference (62F15) Detection theory in information and communication theory (94A13)

Recommendations

• Pruning deep convolutional neural networks architectures with evolution strategy
• Fast convex pruning of deep neural networks
• Structure optimization of convolutional neural networks: a survey
• Sensitivity-informed provable pruning of neural networks
• Survey of deep neural network model compression

Cites work

• A PAC-Bayesian margin bound for linear classifiers
• Tutorial on practical prediction theory for classification

Cited in

(5)

• Structure optimization of convolutional neural networks: a survey
• A reinforcement learning based dynamic room pricing model for hotel industry
• Pruning deep convolutional neural networks architectures with evolution strategy
• Classification of Gaussian spatio-temporal data with stationary separable covariances
• A full stage data augmentation method in deep convolutional neural network for natural image classification

Describes a project that uses

Uses Software

• MobileNets
• ShuffleNet
• Xception
• AlexNet
• ThiNet
• SqueezeNet
• ImageNet
• CIFAR
• Caffe

This page was built for publication: PAC-Bayesian framework based drop-path method for 2D discriminative convolutional network pruning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2192902)