Near Optimal Dimensionality Reductions That Preserve Volumes

  • Avner Magen
  • Anastasios Zouzias
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5171)

Abstract

Let P be a set of n points in Euclidean space and let 0 < ε< 1. A well-known result of Johnson and Lindenstrauss states that there is a projection of P onto a subspace of dimension \(\mathcal{O}(\epsilon^{-2} \log n)\) such that distances change by at most a factor of 1 + ε. We consider an extension of this result. Our goal is to find an analogous dimension reduction where not only pairs but all subsets of at most k points maintain their volume approximately. More precisely, we require that sets of size s ≤ k preserve their volumes within a factor of (1 + ε)s − 1. We show that this can be achieved using \(\mathcal{O}(\max\{\frac{k}{\epsilon},\epsilon^{-2}\log n\})\) dimensions. This in particular means that for \(k = \mathcal{O}(\log n/\epsilon)\) we require no more dimensions (asymptotically) than the special case k = 2, handled by Johnson and Lindenstrauss. Our work improves on a result of Magen (that required as many as \(\mathcal{O}(k\epsilon^{-2}\log n)\) dimensions) and is tight up to a factor of \(\mathcal{O}(1/\epsilon)\). Another outcome of our work is an alternative and greatly simplified proof of the result of Magen showing that all distances between points and affine subspaces spanned by a small number of points are approximately preserved when projecting onto \(\mathcal{O}(k\epsilon^{-2}\log n)\) dimensions.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Achlioptas, D.: Database-friendly random projections. In: PODS 2001: Proceedings of the twentieth ACM SIGMOD-SIGACT-SIGART symposium on Principles of database systems, pp. 274–281. ACM, New York (2001)CrossRefGoogle Scholar
  2. 2.
    Agarwal, P.K., Har-Peled, S., Yu, H.: Embeddings of surfaces, curves, and moving points in euclidean space. In: SCG 2007: Proceedings of the twenty-third annual symposium on Computational geometry, pp. 381–389. ACM, New York (2007)CrossRefGoogle Scholar
  3. 3.
    Ailon, N., Chazelle, B.: Approximate nearest neighbors and the fast johnson-lindenstrauss transform. In: STOC 2006: Proceedings of the thirty-eighth annual ACM symposium on Theory of computing, pp. 557–563. ACM, New York (2006)CrossRefGoogle Scholar
  4. 4.
    Ailon, N., Liberty, E.: Fast dimension reduction using rademacher series on dual bch codes. In: SODA 2008: Proceedings of the nineteenth annual ACM-SIAM symposium on Discrete algorithms, Philadelphia, PA, USA. Society for Industrial and Applied Mathematics, pp. 1–9 (2008)Google Scholar
  5. 5.
    Alon, N.: Problems and results in extremal combinatorics, i. Discrete Math. (273), 31–53 (2003)Google Scholar
  6. 6.
    Clarkson, K.L.: Tighter bounds for random projections of manifolds. In: SCG 2008: Proceedings of the twenty-fourth annual symposium on Computational geometry, pp. 39–48. ACM, New York (2008)CrossRefGoogle Scholar
  7. 7.
    Engebretsen, L., Indyk, P., O’Donnell, R.: Derandomized dimensionality reduction with applications. In: SODA 2002: Proceedings of the thirteenth annual ACM-SIAM symposium on Discrete algorithms, Philadelphia, PA, USA. Society for Industrial and Applied Mathematics, pp. 705–712 (2002)Google Scholar
  8. 8.
    Feige, U.: Approximating the bandwidth via volume respecting embeddings (extended abstract). In: STOC 1998: Proceedings of the thirtieth annual ACM symposium on Theory of computing, pp. 90–99. ACM, New York (1998)CrossRefGoogle Scholar
  9. 9.
    Frankl, P., Maehara, H.: The johnson-lindenstrauss lemma and the sphericity of some graphs. J. Comb. Theory Ser. A 44(3), 355–362 (1987)MathSciNetGoogle Scholar
  10. 10.
    Gordon, L.: Bounds for the distribution of the generalized variance. The Annals of Statistics 17(4), 1684–1692 (1989)MATHCrossRefMathSciNetGoogle Scholar
  11. 11.
    Indyk, P., Naor, A.: Nearest-neighbor-preserving embeddings. ACM Trans. Algorithms 3(3), 31 (2007)CrossRefMathSciNetGoogle Scholar
  12. 12.
    Johnson, W.B., Lindenstrauss, J.: Extensions of lipschitz mappings into a hilbert space. In: Amer. Math. Soc. (ed.) Conference in modern analysis and probability, pp. 189–206. Providence, RI (1984)Google Scholar
  13. 13.
    Liberty, E., Ailon, N., Singer, A.: Fast random projections using lean walsh transforms. In: RANDOM (to appear, 2008)Google Scholar
  14. 14.
    Magen, A.: Dimensionality reductions in ℓ2 that preserve volumes and distance to affine spaces. Discrete & Computational Geometry 38(1), 139–153 (2007)MATHCrossRefMathSciNetGoogle Scholar
  15. 15.
    Matousek, J.: On the variants of johnson lindenstrauss lemma (manuscript) (2006)Google Scholar
  16. 16.
    Prekopa, A.: On random determinants i. Studia Scientiarum Mathematicarum Hungarica (2), 125–132 (1967)Google Scholar
  17. 17.
    Sarlos, T.: Improved approximation algorithms for large matrices via random projections. In: Proceedings of the 47th Annual IEEE Symposium on Foundations of Computer Science, Washington, DC, USA, pp. 143–152. IEEE Computer Society, Los Alamitos (2006)Google Scholar
  18. 18.
    Sivakumar, D.: Algorithmic derandomization via complexity theory. In: STOC 2002: Proceedings of the thirty-fourth annual ACM symposium on Theory of computing, pp. 619–626. ACM, New York (2002)CrossRefGoogle Scholar
  19. 19.
    Vempala, S.: Random projection: A new approach to vlsi layout. In: FOCS 1998: Proceedings of the 39th Annual Symposium on Foundations of Computer Science, Washington, DC, USA, p. 389. IEEE Computer Society, Los Alamitos (1998)Google Scholar
  20. 20.
    Wakin, M.B., Baraniuk, R.G.: Random projections of signal manifolds. In: Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing. ICASSP 2006, vol. 5, p. V (May 2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Avner Magen
    • 1
  • Anastasios Zouzias
    • 1
  1. 1.Department of Computer ScienceUniversity of Toronto 

Personalised recommendations