ML Anthology

Minimal Loss Hashing for Compact Binary Codes

Mohammad Norouzi, David J. Fleet
ICML 2011 pp. 353-360
/icml/2011/norouzi2011icml-minimal/

Abstract

We propose a method for learning similarity-preserving hash functions that map high-dimensional data onto binary codes. The formulation is based on structured prediction with latent variables and a hinge-like loss function. It is efficient to train for large datasets, scales well to large code lengths, and outperforms state-of-the-art methods.

PDF Semantic Scholar

Cite

Text

Norouzi and Fleet. "Minimal Loss Hashing for Compact Binary Codes." International Conference on Machine Learning, 2011.

Markdown

[Norouzi and Fleet. "Minimal Loss Hashing for Compact Binary Codes." International Conference on Machine Learning, 2011.](https://mlanthology.org/icml/2011/norouzi2011icml-minimal/)

BibTeX

@inproceedings{norouzi2011icml-minimal,
  title     = {{Minimal Loss Hashing for Compact Binary Codes}},
  author    = {Norouzi, Mohammad and Fleet, David J.},
  booktitle = {International Conference on Machine Learning},
  year      = {2011},
  pages     = {353-360},
  url       = {https://mlanthology.org/icml/2011/norouzi2011icml-minimal/}
}