Abstract
We consider the problem of encoding a finite set of vectors into a small number of bits while approximately retaining information on the angular distances between the vectors. By deriving improved variance bounds related to binary Gaussian circulant embeddings, we largely fix a gap in the proof of the best known fast binary embedding method. Our bounds also show that well-spreadness assumptions on the data vectors, which were needed in earlier work on variance bounds, are unnecessary. In addition, we propose a new binary embedding with a faster running time on sparse data.
Similar content being viewed by others
References
Ailon, N., Chazelle, B.: The fast Johnson–Lindenstrauss transform and approximate nearest neighbors. SIAM J. Comput. 39(1), 302–322 (2009)
Ailon, N., Liberty, E.: An almost optimal unrestricted fast Johnson–Lindenstrauss transform. ACM Trans. Algorithms 9(3), Art. No. 21 (2013)
Braverman, V., Ostrovsky, R., Rabani, Y.: Rademacher chaos, random Eulerian graphs and the sparse Johnson–Lindenstrauss transform. (2010). arXiv:1011.2590
Cheng, Y., Yu, F.X., Feris, R.S., Kumar, S., Choudhary, A.N., Chang, S.-F.: Fast neural networks with circulant projections. (2015). arXiv:1502.03436
Cheraghchi, M., Guruswami, V., Velingker, A.: Restricted isometry of Fourier matrices and list decodability of random linear codes. SIAM J. Comput. 42(5), 1888–1914 (2013)
Dasgupta, A., Kumar, R., Sarlós, T.: A sparse Johnson–Lindenstrauss transform. In: Proceedings of the 42nd ACM Symposium on Theory of Computing (STOC’10), pp. 341–350. ACM, New York (2010)
Goemans, M.X., Williamson, D.P.: Improved approximation algorithms for maximum cut and satisfiability problems using semidefinite programming. J. Assoc. Comput. Mach. 42(6), 1115–1145 (1995)
Jacques, L., Laska, J.N., Boufounos, P.T., Baraniuk, R.G.: Robust 1-bit compressive sensing via binary stable embeddings of sparse vectors. IEEE Trans. Inform. Theory 59(4), 2082–2102 (2013)
Kane, D.M., Nelson, J.: Sparser Johnson–Lindenstrauss transforms. J. ACM 61(1), Art. No. 4 (2014)
Krahmer, F., Ward, R.: New and improved Johnson–Lindenstrauss embeddings via the restricted isometry property. SIAM J. Math. Anal. 43(3), 1269–1281 (2011)
Oymak, S.: Near-optimal sample complexity bounds for circulant binary embedding. (2016). arXiv:1603.03178
Plan, Y., Vershynin, R.: Robust 1-bit compressed sensing and sparse logistic regression: a convex programming approach. IEEE Trans. Inform. Theory 59(1), 482–494 (2013)
Plan, Y., Vershynin, R.: Dimension reduction by random hyperplane tessellations. Discrete Comput. Geom. 51(2), 438–461 (2014)
Yi, X., Caramanis, C., Price, E.: Binary embedding: fundamental limits and fast algorithm. In: Bach, F., Blei, D. (eds.) Proceedings of the 32nd International Conference on Machine Learning (ICML’15), vol. 37, pp. 2162–2170. JMLR.org (2015)
Yu, F.X., Bhaskara, A., Kumar, S., Gong, Y., Chang, S.-F.: On binary embedding using circulant matrices. (2015). arXiv:1511.06480
Yu, F.X., Kumar, S., Gong, Y., Chang, S.-F.: Circulant binary embedding. In: Proceedings of the 31st International Conference on Machine Learning (ICML’14), vol. 32, pp. 946–954. JMLR.org (2014)
Acknowledgements
The authors would like to thank the reviewers for valuable comments, in particular for a suggestion that substantially shortened the proof of Lemma 2.3. A. Stollenwerk acknowledges funding by the European Research Council through ERC Starting Grant StG 258926. S. Dirksen and A. Stollenwerk acknowledge funding by the Deutsche Forschungsgemeinschaft (DFG) through the project Quantized Compressive Spectrum Sensing (QuaCoSS), which is part of the priority program SPP 1798 Compressed Sensing in Information Processing (COSIP).
Author information
Authors and Affiliations
Corresponding author
Additional information
Editor in Charge: Kenneth Clarkson
Rights and permissions
About this article
Cite this article
Dirksen, S., Stollenwerk, A. Fast Binary Embeddings with Gaussian Circulant Matrices: Improved Bounds. Discrete Comput Geom 60, 599–626 (2018). https://doi.org/10.1007/s00454-017-9964-x
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00454-017-9964-x