Abstract
Neural network pruning techniques can reduce the parameter counts of trained networks by over 90%, decreasing storage requirements and improving computational performance of inference without compromising accuracy. However, contemporary experience is that the sparse architectures produced by pruning are difficult to train from the start, which would similarly improve training performance. We find that a standard pruning technique naturally uncovers subnetworks whose initializations made them capable of training effectively. Based on these results, we articulate the "lottery ticket hypothesis:" 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. The winning tickets we find have won the initialization lottery: their connections have initial weights that make training particularly effective. We present an algorithm to identify winning tickets and a series of experiments that support the lottery ticket hypothesis and the importance of these fortuitous initializations. We consistently find winning tickets that are less than 10-20% of the size of several fully-connected and convolutional feed-forward architectures for MNIST and CIFAR10. Above this size, the winning tickets that we find learn faster than the original network and reach higher test accuracy.
Keywords
PruningInitializationLotteryComputer scienceTicketInferenceMachine learningArtificial intelligenceArtificial neural networkConvolutional neural networkMNIST databaseTest (biology)MathematicsStatisticsComputer security
Affiliated Institutions
Related Publications
Understanding the difficulty of training deep feedforward neural networks
Whereas before 2006 it appears that deep multilayer neural networks were not successfully trained, since then several algorithms have been shown to successfully train them, with...
An empirical evaluation of deep architectures on problems with many factors of variation
Recently, several learning algorithms relying on models with deep architectures have been proposed. Though they have demonstrated impressive performance, to date, they have only...
A Study of Complex Deep Learning Networks on High-Performance, Neuromorphic, and Quantum Computers
Current deep learning approaches have been very successful using convolutional neural networks trained on large graphical-processing-unit-based computers. Three limitations of t...
Deep Sparse Rectifier Neural Networks
While logistic sigmoid neurons are more biologically plausible than hyperbolic tangent neurons, the latter work better for training multi-layer neural networks. This paper shows...
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. T...
Publication Info
- Year
- 2018
- Type
- preprint
- Citations
- 1317
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
1317
OpenAlex
Cite This
Jonathan Frankle,
Michael Carbin
(2018).
The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks.
arXiv (Cornell University)
.
https://doi.org/10.48550/arxiv.1803.03635
Identifiers
- DOI
- 10.48550/arxiv.1803.03635