Improving Sign Random Projections With Additional Information

Keegan Kang, Weipin Wong
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:2479-2487, 2018.

Abstract

Sign random projections (SRP) is a technique which allows the user to quickly estimate the angular similarity and inner products between data. We propose using additional information to improve these estimates which is easy to implement and cost efficient. We prove that the variance of our estimator is lower than the variance of SRP. Our proposed method can also be used together with other modifications of SRP, such as Super-Bit LSH (SBLSH). We demonstrate the effectiveness of our method on the MNIST test dataset and the Gisette dataset. We discuss how our proposed method can be extended to random projections or even other hashing algorithms.

Cite this Paper

BibTeX
@InProceedings{pmlr-v80-kang18b, title = {Improving Sign Random Projections With Additional Information}, author = {Kang, Keegan and Wong, Weipin}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {2479--2487}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/kang18b/kang18b.pdf}, url = {https://proceedings.mlr.press/v80/kang18b.html}, abstract = {Sign random projections (SRP) is a technique which allows the user to quickly estimate the angular similarity and inner products between data. We propose using additional information to improve these estimates which is easy to implement and cost efficient. We prove that the variance of our estimator is lower than the variance of SRP. Our proposed method can also be used together with other modifications of SRP, such as Super-Bit LSH (SBLSH). We demonstrate the effectiveness of our method on the MNIST test dataset and the Gisette dataset. We discuss how our proposed method can be extended to random projections or even other hashing algorithms.} }
Endnote
%0 Conference Paper %T Improving Sign Random Projections With Additional Information %A Keegan Kang %A Weipin Wong %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-kang18b %I PMLR %P 2479--2487 %U https://proceedings.mlr.press/v80/kang18b.html %V 80 %X Sign random projections (SRP) is a technique which allows the user to quickly estimate the angular similarity and inner products between data. We propose using additional information to improve these estimates which is easy to implement and cost efficient. We prove that the variance of our estimator is lower than the variance of SRP. Our proposed method can also be used together with other modifications of SRP, such as Super-Bit LSH (SBLSH). We demonstrate the effectiveness of our method on the MNIST test dataset and the Gisette dataset. We discuss how our proposed method can be extended to random projections or even other hashing algorithms.
APA
Kang, K. & Wong, W.. (2018). Improving Sign Random Projections With Additional Information. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:2479-2487 Available from https://proceedings.mlr.press/v80/kang18b.html.

Related Material