Optimized intrinsic dimension estimation using nearest neighbor graphs (original) (raw)

Abstract

We develop an approach to intrinsic dimension estimation based on k-nearest neighbor (kNN) distances. The dimension estimator is derived using a general theory on functionals of kNN density estimates. This enables us to predict the performance of the dimension estimation algorithm. In addition, it allows for optimization of free parameters in the algorithm. We validate our theory through simulations and compare our estimator to previous kNN based dimensionality estimation approaches.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (8)

  1. REFERENCES
  2. J.A. Costa, A. Girotra, and A.O. Hero, "Estimating local in- trinsic dimension with k-nearest neighbor graphs," in 2005 IEEE/SP 13th Workshop on Statistical Signal Processing, 2005, pp. 417-422.
  3. A.M. Farahmand, C. Sepesvari, and J-Y Audibert, "Manifold- adaptive dimension estimation," Proc of 24th Intl Conf on Ma- chine Learning, pp. 265-272, 2007.
  4. N. Leonenko, L. Prozanto, and V. Savani, "A class of rényi information estimators for multidimensional densities," Annals of Statistics, vol. 36, pp. 2153-2182, 2008.
  5. K. Fukunaga and L. Hostetler, "Optimization of k nearest neigh- bor density estimates," IEEE Transactions on Information The- ory, vol. 19, no. 3, pp. 320-326, 1973.
  6. K. Sricharan, R. Raich, and A. O. Hero, "Plug-in estimators for non-linear functionals of densities," Technical Report, Commu- nications and Signal Processing Laboratory, The University of Michigan, September 2009, (To appear).
  7. K. Sricharan, R. Raich, and A. O. Hero, "Global performance prediction for divergence-based image registration criteria," in Proc. IEEE Workshop on Statistical Signal Processing, 2009, (To appear).
  8. V. C. Raykar and R. Duraiswami, "Fast optimal bandwidth se- lection for kernel density estimation," in Proceedings of the sixth SIAM International Conference on Data Mining, J. Ghosh, D. Lambert, D. Skillicorn, and J. Srivastava, Eds., 2006, pp. 524-528.