Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations

Abstract

We present a new approach to learn compressible representations in deep architectures with an end-to-end training strategy. Our method is based on a soft (continuous) relaxation of quantization and entropy, which we anneal to their discrete counterparts throughout training. We showcase this method for two challenging applications: Image compression and neural network compression. While these tasks have typically been approached with different methods, our soft-to-hard quantization approach gives results competitive with the state-of-the-art for both.

Cite

Text

Agustsson et al. "Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations." Neural Information Processing Systems, 2017.

Markdown

[Agustsson et al. "Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations." Neural Information Processing Systems, 2017.](https://mlanthology.org/neurips/2017/agustsson2017neurips-softtohard/)

BibTeX

@inproceedings{agustsson2017neurips-softtohard,
  title     = {{Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations}},
  author    = {Agustsson, Eirikur and Mentzer, Fabian and Tschannen, Michael and Cavigelli, Lukas and Timofte, Radu and Benini, Luca and Gool, Luc V.},
  booktitle = {Neural Information Processing Systems},
  year      = {2017},
  pages     = {1141-1151},
  url       = {https://mlanthology.org/neurips/2017/agustsson2017neurips-softtohard/}
}