Nonlinear function approximation over high-dimensional domains

Details Nonlinear function approximation over high-dimensional domains Nonlinear function approximation over high-dimensional domains

2007-09-05

2011-10-25

Jacob, Mary (Department: 2123)

Data processing: structural design, modeling, simulation, and em

Modeling by mathematical expression

C702S183000, C714S047300

Reexamination Certificate

active

08046200

ABSTRACT:
An algorithm is disclosed for constructing nonlinear models from high-dimensional scattered data. The algorithm progresses iteratively adding a new basis function at each step to refine the model. The placement of the basis functions is driven by a statistical hypothesis test that reveals geometric structure when it fails. At each step the added function is fit to data contained in a spatio-temporally defined local region to determine the parameters, in particular, the scale of the local model. The proposed method requires no ad hoc parameters. Thus, the number of basis functions required for an accurate fit is determined automatically by the algorithm. The approach may be applied to problems including modeling data on manifolds and the prediction of financial time-series. The algorithm is presented in the context of radial basis functions but in principle can be employed with other methods for function approximation such as multi-layer perceptrons.

REFERENCES:
patent: 4752957 (1988-06-01), Maeda
patent: 5223207 (1993-06-01), Gross et al.
patent: 5377306 (1994-12-01), Broomhead et al.
patent: 5453940 (1995-09-01), Broomhead et al.
patent: 5475793 (1995-12-01), Broomhead et al.
patent: 5493516 (1996-02-01), Broomhead et al.
patent: 5566002 (1996-10-01), Shikakura
patent: 5586066 (1996-12-01), White et al.
patent: 5629872 (1997-05-01), Gross et al.
patent: 5667837 (1997-09-01), Broomhead et al.
patent: 5731989 (1998-03-01), Tenny et al.
patent: 5745382 (1998-04-01), Vilim et al.
patent: 5761090 (1998-06-01), Gross et al.
patent: 5764509 (1998-06-01), Gross et al.
patent: 5835682 (1998-11-01), Broomhead et al.
patent: 5842194 (1998-11-01), Arbuckle
patent: 5845123 (1998-12-01), Johnson et al.
patent: 6118549 (2000-09-01), Katougi et al.
patent: 6119111 (2000-09-01), Gross et al.
patent: 6131076 (2000-10-01), Stephan et al.
patent: RE37488 (2001-12-01), Broomhead et al.
patent: 6466685 (2002-10-01), Fukui et al.
patent: 6493465 (2002-12-01), Mori et al.
patent: 6625569 (2003-09-01), James et al.
patent: 6636628 (2003-10-01), Wang et al.
patent: 6636862 (2003-10-01), Lundahl et al.
patent: 6751348 (2004-06-01), Buzuloiu et al.
patent: 6754675 (2004-06-01), Abdel-Mottaleb et al.
patent: 6758574 (2004-07-01), Roberts
patent: 6775417 (2004-08-01), Hong et al.
patent: 6798539 (2004-09-01), Wang et al.
patent: 6873730 (2005-03-01), Chen
patent: 6996257 (2006-02-01), Wang
patent: 7023577 (2006-04-01), Watanabe et al.
patent: 7068838 (2006-06-01), Manbeck et al.
patent: 7080290 (2006-07-01), James et al.
patent: 7142697 (2006-11-01), Huang et al.
patent: 7224835 (2007-05-01), Maeda et al.
patent: 7262881 (2007-08-01), Livens et al.
patent: 2002/0010691 (2002-01-01), Chen
patent: 2002/0140985 (2002-10-01), Hudson
patent: 2002/0191034 (2002-12-01), Sowizral et al.
patent: 2003/0065409 (2003-04-01), Raeth et al.
patent: 2004/0130546 (2004-07-01), Porikli
patent: 2007/0097965 (2007-05-01), Qiao et al.
patent: 2008/0175446 (2008-07-01), Kirby et al.
patent: 2008/0256130 (2008-10-01), Kirby et al.
patent: WO 02/095534 (2002-11-01), None
“Chapter 4 Clipping a Gaussian Process”, date unknown, pp. 33-43.
“Comparing Model Structures” Model Structure and Model Validation, date unknown, pp. 498-519.
“Estimation of the Mean and the Autocovariance Function”, Time series: Theory and Methods, 1991, pp. 218-237.
Ahmed et al., “Adaptive RBF Neural Network in Signal Detection”, In 1994 IEEE International Symposium on Circuits and Systems, ISCAS, vol. 6, pp. 265-268, May-Jun. 1994.
Aiyar et al., “Minimal Resource Allocation Network for Call Admission control (CAC) of ATM Networks”, IEEE, 2000, p. 498.
Akaike, “A New Look at the Statistical Model Identificaiton”, IEEE Transactions on Automatic Control, Dec. 1974, vol. AC-19, No. 6, pp. 716-723.
Allasia “Approximating Potential Integrals by Cardinal Basis Interpolants on Multivariate Scattered Data”, Computers and Mathematics with Applications 43 (2002), pp. 275-287.
Anderle “Modeling Geometric Structure in Noisy Data”, Dissertation for Colorado State University, Summer 2001, 117 pages.
Anderle et al., “Correlation Feedback Resource Allocation RBF”, IEEE, 2001, pp. 1949-1953.
Andrieu et al., “Robust Full Bayesian Learning for Radial Basis Networks”, Neural Computation, 2001, vol. 13, pp. 2359-2407.
Arellano-Valle et al., “On fundamental skew distributions”, Journal of Multivariate Analysis, 2005, vol. 96, pp. 93-116.
Arnold et al., “The skew-Cauchy distribution”, Statistics & Probability Letters, 2000, vol. 49, pp. 285-290.
Azam et al., “An Alternate Radial Basis Function Neural Network Model”, in 2000 IEEE International Conference on Systems, Man, and Cybernetics, vol. 4, pp. 2679-2684, Oct. 8-11, 2000.
Azzalini “Further Results on a Class of Distributions which Includes the Normal Ones”, Statistica, 1986, anno XLVI, No. 2, pp. 199-208.
Azzalini “Statistical applications of the multivariate skew normal distribution”, J. R. Statist. Soc. B, 1999, vol. 61, Part 3, pp. 579-602.
Azzalini et al., “A Class of Distributions which Includes the Normal Ones”, Scand J Statist, 1985, vol. 12, pp. 171-178.
Azzalini et al., “Distributions generated by perturbation of symmetry with emphasis on a multivariate skew t distribution”, Journal of Royal Statistical Society, 65(B) :367-389, 2003.
Azzalini et al., “The Multivariate Skew-Normal Distribution”, Biometrika, Dec. 1996, vol. 83, No. 4, pp. 715-726.
Balakrishnan et al., “An Operator Splitting-Radial Basis Function method for the Solution of Transient Nonlinear Poisson Problems”, Computers and Mathematics with Applications 43 (2002), pp. 289-304.
Ball et al., “Eigenvalues of Euclidean Distance Matrices”, Journal of Approximation Theory, 1992, vol. 68, pp. 74-82.
Ball et al., “On the Sensitivity of Radial Basis Interpolation to Minimal Data Separation Distance”, Constr. Approx., 1992, vol. 8, pp. 401-426.
Barnett et al., “Zero-Crossing Rates of Functions of Gaussian Processes”, IEEE Transactions on Information Theory, Jul. 1991, vol. 37, No. 4, pp. 1188-1194.
Barnett et al., “Zero-Crossing Rates of Mixtures and Products of Gaussian Processes”, IEEE Transactions on Information Theory, Jul. 1998, vol. 44, No. 4, pp. 1672-1677.
Baxter et al., “Preconditioned Conjugate Gradients, Radial Basis Functions, and Toeplitz Matrices”, Computers and Mathematics with Applications 43 (2002), pp. 305-318.
Behboodian et al., “A new class of skew-Cauchy distributions”, Statistics & Probability Letters, 2006, vol. 76, pp. 1488-1493.
Behrens et al., “Grid-Free Adaptive Semi-Lagrangian Advection Using Radial Basis Functions”, Computers and Mathematics with Applications 43 (2002), pp. 319-327.
Belytschko et al., “Stability Analysis of Particle Methods with Corrected Derivatives”, Computers and Mathematics with Applications 43 (2002), pp. 329-350.
Bishop, “Radial Basis Functions”, Neural Networks for Pattern Recognition, Oxford University Press, Oxford, U.K., 1995, pp. 164-193.
Blachman “Zero-Crossing Rate for the Sum of Two Sinusoids or a Signal Plus Noise”, IEEE Transactions on Information Theory, Nov. 1975, pp. 671-675.
Bozzini et al., “Interpolation by basis functions of different scales and shapes”, Calcolo, 2004, vol. 41, pp. 77-87.
Bozzini et al., “Interpolation by basis functions of different scales and shapes”, date unknown, pp. 1-13.
Branco et al., “A General Class of Multivariate Skew-Elliptical Distributions”, Journal of Multivariate Analysis, 2001, vol. 79, pp. 99-113.
Brand, “Charting a manifold”, Oct. 2002, 8 pages.
Brand, “Continuous nonlinear dimensionality reduction by kernel eigenmaps”, MERL, Apr. 2003, 8 pages.
Brass et al., “Hybrid Monte Carl

Affiliated with

Also associated with

Rating

Nonlinear function approximation over high-dimensional domains does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Nonlinear function approximation over high-dimensional domains, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Nonlinear function approximation over high-dimensional domains will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-4265836