Corpus ID: 15582471

A Deep Neural Network Compression Pipeline: Pruning, Quantization, Huffman Encoding

@inproceedings{Han2015ADN,
  title={A Deep Neural Network Compression Pipeline: Pruning, Quantization, Huffman Encoding},
  author={Song Han and Huizi Mao and W. J. Dally},
  year={2015}
}
  • Song Han, Huizi Mao, W. J. Dally
  • Published 2015
  • Computer Science
  • Neural networks are both computationally intensive and memory intensive, making them difficult to deploy on embedded systems with limited hardware resources. [...] Key Method Our method first prunes the network by learning only the important connections. Next, we quantize the weights to enforce weight sharing, finally, we apply Huffman encoding. After the first two steps we retrain the network to fine tune the remaining connections and the quantized centroids.Expand Abstract
    49 Citations
    Deep Neural Network Compression Method Based on Product Quantization
    Compact Deep Convolutional Neural Networks With Coarse Pruning
    • 38
    • PDF
    Scalpel: Customizing DNN pruning to the underlying hardware parallelism
    • 215
    • Highly Influenced
    • PDF
    Structured Pruning of Deep Convolutional Neural Networks
    • 337
    • Highly Influenced
    • PDF
    Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applications
    • 523
    • Highly Influenced
    • PDF
    Pruning Filters for Efficient ConvNets
    • 1,411
    • Highly Influenced
    • PDF
    Compressing Convolutional Neural Networks in the Frequency Domain
    • 60
    • PDF
    Deep Neural Network Approximation using Tensor Sketching
    • 6
    • PDF
    Pruning Deep Convolutional Neural Networks for Fast Inference
    • Highly Influenced
    • PDF
    SEP-Nets: Small and Effective Pattern Networks
    • 10
    • PDF

    References

    SHOWING 1-10 OF 24 REFERENCES
    Learning both Weights and Connections for Efficient Neural Network
    • 2,884
    • PDF
    Memory Bounded Deep Convolutional Networks
    • 127
    • PDF
    Improving the speed of neural networks on CPUs
    • 592
    • PDF
    ImageNet classification with deep convolutional neural networks
    • 59,744
    • PDF
    Going deeper with convolutions
    • 22,376
    • PDF
    Deep Fried Convnets
    • 209
    • PDF