methodology-9
Network Pruning
3260 papers • 126 benchmarks • 313 datasets
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
(Image credit: Papersgraph)
Benchmarks
These leaderboards are used to track progress in network-pruning-9
Trend
Dataset
Best Model
Actions
ImageNet
ImageNet
ImageNet - ResNet 50 - 90% sparsity
ImageNet - ResNet 50 - 90% sparsity
CIFAR-100
Libraries
iUse these libraries to find network-pruning-9 models and implementations
4 papers 43
Datasets
Subtasks
No subtasks available.
Most implemented papers
The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
Jonathan Frankle, Michael Carbin•Thu Mar 08 2018
This work finds that dense, randomly-initialized, feed-forward networks contain subnetworks ("winning tickets") that - when trained in isolation - reach test accuracy comparable to the original network in a similar number of iterations, and articulate the "lottery ticket hypothesis".
3981
