Michael David Kirby

US Patent:

7917540, Mar 29, 2011

Filed:

Feb 22, 2008

Appl. No.:

12/036117

Inventors:

Michael Joseph Kirby - Fort Collins CO, US
Christopher Scott Peterson - Fort Collins CO, US

Assignee:

Colorado State University Research Foundation - Fort Collins CO

International Classification:

G06F 7/00
G06K 9/54

US Classification:

707802, 382170, 382181

Abstract:

Variations in the states of patterns can be exploited for their discriminatory information and should not be discarded as noise. A pattern recognition system compares a data set of unlabeled patterns having variations of state in a set-by-set comparison with labeled arrays of individual data sets of multiple patterns also having variations of state. The individual data sets are each mapped to a point on a parameter space, and the points of each labeled array define a subset of the parameter space. If the point associated with the data set of unlabeled patterns satisfies a similarity criterion on the parameter space subset of a labeled array, the data set of unlabeled patterns is assigned to the class attributed to that labeled array.

US Patent:

7921146, Apr 5, 2011

Filed:

Nov 1, 2005

Appl. No.:

11/263740

Inventors:

Yue Qiao - Longmont CO, US
Michael Joseph Kirby - Fort Collins CO, US

Assignee:

Infoprint Solutions Company, LLC - Boulder CO

International Classification:

G06F 7/38
H04N 1/46

US Classification:

708290, 358525

Abstract:

An apparatus, system, and method are disclosed for interpolating data. A cluster module locates a plurality of variable granularity data clusters within a source data set using a center selection/clustering algorithm. A conversion selection module selects a conversion function for converting each data cluster. A norm selection module selects a norm function for each data cluster. An interpolation module converts each data cluster. In one embodiment, a cross validation module iteratively cross validates and optimizes the data conversions.

US Patent:

8116566, Feb 14, 2012

Filed:

Aug 28, 2007

Appl. No.:

11/846486

Inventors:

Michael Joseph Kirby - Fort Collins CO, US
James Ross Beveridge - Fort Collins CO, US
Jen-Mei Chang - Fort Collins CO, US
Bruce Anthony Draper - Fort Collins CO, US
Holger Philipp Kley - Fort Collins CO, US
Christopher Scott Peterson - Fort Collins CO, US

Assignee:

Colorado State University Research Foundation - Fort Collins CO

International Classification:

G06K 9/00

US Classification:

382181, 382118

Abstract:

A pattern recognition system compares a set of unlabeled images (or other patterns) having a variation of state in a set-by-set comparison with individual data sets of multiple labeled images (or other patterns) also having a variation of state. The individual data sets are each mapped to a point on a parameter space (e. g. , a Grassmannian manifold, a Stiefel manifold, a flag manifold, etc. ), and the set of unlabeled images is mapped to a point in the same parameter space. If the point associated with the set of unlabeled images satisfies a distance criterion on the parameter space with regard to one of the points on the parameter space, the data set of unlabeled images is assigned to the class attributed to that point.

US Patent:

8521488, Aug 27, 2013

Filed:

Jun 10, 2011

Appl. No.:

13/134597

Inventors:

Michael J. Kirby - Fort Collins CO, US
Arthur A. Jamshidi - Plainsboro NJ, US

Assignee:

National Science Foundation - Arlington VA

International Classification:

G06F 17/10

US Classification:

703 2, 702183, 714 47

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.

US Patent:

8046200, Oct 25, 2011

Filed:

Sep 5, 2007

Appl. No.:

11/899625

Inventors:

Michael J. Kirby - Fort Collins CO, US
Arthur A. Jamshidi - Fort Collins CO, US

Assignee:

Colorado State University Research Foundation - Fort Collins CO

International Classification:

G06F 7/60
G06F 17/10

US Classification:

703 2, 702183, 714 47

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.