Circulant Binary Embedding

Felix Yu, Sanjiv Kumar, Yunchao Gong, Shih-Fu Chang
Proceedings of the 31st International Conference on Machine Learning, PMLR 32(2):946-954, 2014.

Abstract

Binary embedding of high-dimensional data requires long codes to preserve the discriminative power of the input space. Traditional binary coding methods often suffer from very high computation and storage costs in such a scenario. To address this problem, we propose Circulant Binary Embedding (CBE) which generates binary codes by projecting the data with a circulant matrix. The circulant structure enables the use of Fast Fourier Transformation to speed up the computation. Compared to methods that use unstructured matrices, the proposed method improves the time complexity from \mathcalO(d^2) to \mathcalO(d\logd), and the space complexity from \mathcalO(d^2) to \mathcalO(d) where d is the input dimensionality. We also propose a novel time-frequency alternating optimization to learn data-dependent circulant projections, which alternatively minimizes the objective in original and Fourier domains. We show by extensive experiments that the proposed approach gives much better performance than the state-of-the-art approaches for fixed time, and provides much faster computation with no performance degradation for fixed number of bits.

Cite this Paper


BibTeX
@InProceedings{pmlr-v32-yub14, title = {Circulant Binary Embedding}, author = {Yu, Felix and Kumar, Sanjiv and Gong, Yunchao and Chang, Shih-Fu}, booktitle = {Proceedings of the 31st International Conference on Machine Learning}, pages = {946--954}, year = {2014}, editor = {Xing, Eric P. and Jebara, Tony}, volume = {32}, number = {2}, series = {Proceedings of Machine Learning Research}, address = {Bejing, China}, month = {22--24 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v32/yub14.pdf}, url = {https://proceedings.mlr.press/v32/yub14.html}, abstract = {Binary embedding of high-dimensional data requires long codes to preserve the discriminative power of the input space. Traditional binary coding methods often suffer from very high computation and storage costs in such a scenario. To address this problem, we propose Circulant Binary Embedding (CBE) which generates binary codes by projecting the data with a circulant matrix. The circulant structure enables the use of Fast Fourier Transformation to speed up the computation. Compared to methods that use unstructured matrices, the proposed method improves the time complexity from \mathcalO(d^2) to \mathcalO(d\logd), and the space complexity from \mathcalO(d^2) to \mathcalO(d) where d is the input dimensionality. We also propose a novel time-frequency alternating optimization to learn data-dependent circulant projections, which alternatively minimizes the objective in original and Fourier domains. We show by extensive experiments that the proposed approach gives much better performance than the state-of-the-art approaches for fixed time, and provides much faster computation with no performance degradation for fixed number of bits.} }
Endnote
%0 Conference Paper %T Circulant Binary Embedding %A Felix Yu %A Sanjiv Kumar %A Yunchao Gong %A Shih-Fu Chang %B Proceedings of the 31st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2014 %E Eric P. Xing %E Tony Jebara %F pmlr-v32-yub14 %I PMLR %P 946--954 %U https://proceedings.mlr.press/v32/yub14.html %V 32 %N 2 %X Binary embedding of high-dimensional data requires long codes to preserve the discriminative power of the input space. Traditional binary coding methods often suffer from very high computation and storage costs in such a scenario. To address this problem, we propose Circulant Binary Embedding (CBE) which generates binary codes by projecting the data with a circulant matrix. The circulant structure enables the use of Fast Fourier Transformation to speed up the computation. Compared to methods that use unstructured matrices, the proposed method improves the time complexity from \mathcalO(d^2) to \mathcalO(d\logd), and the space complexity from \mathcalO(d^2) to \mathcalO(d) where d is the input dimensionality. We also propose a novel time-frequency alternating optimization to learn data-dependent circulant projections, which alternatively minimizes the objective in original and Fourier domains. We show by extensive experiments that the proposed approach gives much better performance than the state-of-the-art approaches for fixed time, and provides much faster computation with no performance degradation for fixed number of bits.
RIS
TY - CPAPER TI - Circulant Binary Embedding AU - Felix Yu AU - Sanjiv Kumar AU - Yunchao Gong AU - Shih-Fu Chang BT - Proceedings of the 31st International Conference on Machine Learning DA - 2014/06/18 ED - Eric P. Xing ED - Tony Jebara ID - pmlr-v32-yub14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 32 IS - 2 SP - 946 EP - 954 L1 - http://proceedings.mlr.press/v32/yub14.pdf UR - https://proceedings.mlr.press/v32/yub14.html AB - Binary embedding of high-dimensional data requires long codes to preserve the discriminative power of the input space. Traditional binary coding methods often suffer from very high computation and storage costs in such a scenario. To address this problem, we propose Circulant Binary Embedding (CBE) which generates binary codes by projecting the data with a circulant matrix. The circulant structure enables the use of Fast Fourier Transformation to speed up the computation. Compared to methods that use unstructured matrices, the proposed method improves the time complexity from \mathcalO(d^2) to \mathcalO(d\logd), and the space complexity from \mathcalO(d^2) to \mathcalO(d) where d is the input dimensionality. We also propose a novel time-frequency alternating optimization to learn data-dependent circulant projections, which alternatively minimizes the objective in original and Fourier domains. We show by extensive experiments that the proposed approach gives much better performance than the state-of-the-art approaches for fixed time, and provides much faster computation with no performance degradation for fixed number of bits. ER -
APA
Yu, F., Kumar, S., Gong, Y. & Chang, S.. (2014). Circulant Binary Embedding. Proceedings of the 31st International Conference on Machine Learning, in Proceedings of Machine Learning Research 32(2):946-954 Available from https://proceedings.mlr.press/v32/yub14.html.

Related Material