SOTAVerified

Network Pruning

Network Pruning is a popular approach to reduce a heavy network to obtain a light-weight form by removing redundancy in the heavy network. In this approach, a complex over-parameterized network is first trained, then pruned based on come criterions, and finally fine-tuned to achieve comparable performance with reduced parameters.

Source: Ensemble Knowledge Distillation for Learning Improved and Efficient Networks

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 101–125 of 534 papers

TitleStatusHype
Dynamic ASR Pathways: An Adaptive Masking Approach Towards Efficient Pruning of A Multilingual ASR Model—0
Unveiling Invariances via Neural Network Pruning—0
Ensemble Mask Networks—0
EDAC: Efficient Deployment of Audio Classification Models For COVID-19 DetectionCode0
Adaptive Consensus: A network pruning approach for decentralized optimization—0
Efficient and Flexible Neural Network Training through Layer-wise Feedback PropagationCode1
To prune or not to prune : A chaos-causality approach to principled pruning of dense neural networks—0
A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and RecommendationsCode2
Pruning a neural network using Bayesian inference—0
Accurate Neural Network Pruning Requires Rethinking Sparse Optimization—0
Model Compression Methods for YOLOv5: A Review—0
SqueezerFaceNet: Reducing a Small Face Recognition CNN Even More Via Filter Pruning—0
Neural Network Pruning as Spectrum Preserving Process—0
Distilled Pruning: Using Synthetic Data to Win the LotteryCode0
Pruning vs Quantization: Which is Better?Code1
Structured Network Pruning by Measuring Filter-wise Interactions—0
Filter Pruning for Efficient CNNs via Knowledge-driven Differential Filter SamplerCode0
Low-Rank Prune-And-Factorize for Language Model Compression—0
Neural Network Pruning for Real-time Polyp Segmentation—0
A Simple and Effective Pruning Approach for Large Language ModelsCode2
Representation and decomposition of functions in DAG-DNNs and structural network pruning—0
Improving Generalization in Meta-Learning via Meta-Gradient AugmentationCode0
Resource Efficient Neural Networks Using Hessian Based Pruning—0
How Sparse Can We Prune A Deep Network: A Fundamental Limit ViewpointCode1
Does a sparse ReLU network training problem always admit an optimum?—0
Show:102550
← PrevPage 5 of 22Next →
All datasetsImageNetImageNet - ResNet 50 - 90% sparsityCIFAR-100CIFAR-10MNIST

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1ResNet50-2.3 GFLOPsAccuracy78.79—Unverified
2ResNet50-1.5 GFLOPsAccuracy78.07—Unverified
3ResNet50 2.5 GFLOPSAccuracy78—Unverified
4RegX-1.6GAccuracy77.97—Unverified
5ResNet50 2.0 GFLOPSAccuracy77.7—Unverified
6ResNet50-3G FLOPsAccuracy77.1—Unverified
7ResNet50-2G FLOPsAccuracy76.4—Unverified
8ResNet50-1G FLOPsAccuracy76.38—Unverified
9TAS-pruned ResNet-50Accuracy76.2—Unverified
10ResNet50Accuracy75.59—Unverified
#ModelMetricClaimedVerifiedStatus
1FeatherTop-1 Accuracy76.93—Unverified
2SpartanTop-1 Accuracy76.17—Unverified
3ST-3Top-1 Accuracy76.03—Unverified
4AC/DCTop-1 Accuracy75.64—Unverified
5CSTop-1 Accuracy75.5—Unverified
6ProbMaskTop-1 Accuracy74.68—Unverified
7STRTop-1 Accuracy74.31—Unverified
8DNWTop-1 Accuracy74—Unverified
9GMPTop-1 Accuracy73.91—Unverified
#ModelMetricClaimedVerifiedStatus
1+U-DML*Inference Time (ms)675.56—Unverified
2DenseAccuracy79—Unverified
3AC/DCAccuracy78.2—Unverified
4Beta-RankAccuracy74.01—Unverified
5TAS-pruned ResNet-110Accuracy73.16—Unverified
#ModelMetricClaimedVerifiedStatus
1TAS-pruned ResNet-110Accuracy94.33—Unverified
2ShuffleNet – QuantisedInference Time (ms)23.15—Unverified
3AlexNet – QuantisedInference Time (ms)5.23—Unverified
4MobileNet – QuantisedInference Time (ms)4.74—Unverified
#ModelMetricClaimedVerifiedStatus
1FFN-ShapleyPrunedAvg #Steps12.05—Unverified