Sorry, you need to enable JavaScript to visit this website.

Graphical and kernel methods (MLR-GRKN)

Semi-supervised learning in the presence of novel class instances

Paper Details

Authors:
Anh T. Pham, Raviv Raich, Xiaoli Z. Fern
Submitted On:
19 March 2016 - 8:25pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

Poster_ICASSP.pdf

(130 downloads)

Keywords

Subscribe

[1] Anh T. Pham, Raviv Raich, Xiaoli Z. Fern, "Semi-supervised learning in the presence of novel class instances", IEEE SigPort, 2016. [Online]. Available: http://sigport.org/838. Accessed: Apr. 23, 2017.
@article{838-16,
url = {http://sigport.org/838},
author = {Anh T. Pham; Raviv Raich; Xiaoli Z. Fern },
publisher = {IEEE SigPort},
title = {Semi-supervised learning in the presence of novel class instances},
year = {2016} }
TY - EJOUR
T1 - Semi-supervised learning in the presence of novel class instances
AU - Anh T. Pham; Raviv Raich; Xiaoli Z. Fern
PY - 2016
PB - IEEE SigPort
UR - http://sigport.org/838
ER -
Anh T. Pham, Raviv Raich, Xiaoli Z. Fern. (2016). Semi-supervised learning in the presence of novel class instances. IEEE SigPort. http://sigport.org/838
Anh T. Pham, Raviv Raich, Xiaoli Z. Fern, 2016. Semi-supervised learning in the presence of novel class instances. Available at: http://sigport.org/838.
Anh T. Pham, Raviv Raich, Xiaoli Z. Fern. (2016). "Semi-supervised learning in the presence of novel class instances." Web.
1. Anh T. Pham, Raviv Raich, Xiaoli Z. Fern. Semi-supervised learning in the presence of novel class instances [Internet]. IEEE SigPort; 2016. Available from : http://sigport.org/838

Models for Spectral Clustering and Their Applications


Microarray Spectral Clustering.

Ph.D. Thesis by Donald McCuan (advisor Andrew Knyazev), Department of Mathematical and Statistical Sciences, University of Colorado Denver, 2012, originally posted at http://math.ucdenver.edu/theses/McCuan_PhdThesis.pdf (392 downloads)

Paper Details

Authors:
Donald Donald
Submitted On:
23 February 2016 - 1:44pm
Short Link:
Type:
Document Year:
Cite

Document Files

McCuan_PhdThesis.pdf

(392 downloads)

Keywords

Subscribe

[1] Donald Donald, "Models for Spectral Clustering and Their Applications", IEEE SigPort, 2015. [Online]. Available: http://sigport.org/564. Accessed: Apr. 23, 2017.
@article{564-15,
url = {http://sigport.org/564},
author = {Donald Donald },
publisher = {IEEE SigPort},
title = {Models for Spectral Clustering and Their Applications},
year = {2015} }
TY - EJOUR
T1 - Models for Spectral Clustering and Their Applications
AU - Donald Donald
PY - 2015
PB - IEEE SigPort
UR - http://sigport.org/564
ER -
Donald Donald. (2015). Models for Spectral Clustering and Their Applications. IEEE SigPort. http://sigport.org/564
Donald Donald, 2015. Models for Spectral Clustering and Their Applications. Available at: http://sigport.org/564.
Donald Donald. (2015). "Models for Spectral Clustering and Their Applications." Web.
1. Donald Donald. Models for Spectral Clustering and Their Applications [Internet]. IEEE SigPort; 2015. Available from : http://sigport.org/564

Multigrid Eigensolvers for Image Segmentation


Spectral Image Segmentation

Presentation at LANL and UC Davis, 2009. Originally posted at http://math.ucdenver.edu/~aknyazev/research/conf/

LANL09.ppt

Office presentation icon LANL09.ppt (211 downloads)

Paper Details

Authors:
Submitted On:
23 February 2016 - 1:44pm
Short Link:
Type:
Document Year:
Cite

Document Files

LANL09.ppt

(211 downloads)

Keywords

Subscribe

[1] , "Multigrid Eigensolvers for Image Segmentation", IEEE SigPort, 2015. [Online]. Available: http://sigport.org/562. Accessed: Apr. 23, 2017.
@article{562-15,
url = {http://sigport.org/562},
author = { },
publisher = {IEEE SigPort},
title = {Multigrid Eigensolvers for Image Segmentation},
year = {2015} }
TY - EJOUR
T1 - Multigrid Eigensolvers for Image Segmentation
AU -
PY - 2015
PB - IEEE SigPort
UR - http://sigport.org/562
ER -
. (2015). Multigrid Eigensolvers for Image Segmentation. IEEE SigPort. http://sigport.org/562
, 2015. Multigrid Eigensolvers for Image Segmentation. Available at: http://sigport.org/562.
. (2015). "Multigrid Eigensolvers for Image Segmentation." Web.
1. . Multigrid Eigensolvers for Image Segmentation [Internet]. IEEE SigPort; 2015. Available from : http://sigport.org/562

Novel data clustering for microarrays and image segmentation


Spectral Clustering Eigenvalue Problem

We develop novel algorithms and software on parallel computers for data clustering of large datasets. We are interested in applying our approach, e.g., for analysis of large datasets of microarrays or tiling arrays in molecular biology and for segmentation of high resolution images.

Paper Details

Authors:
Submitted On:
23 February 2016 - 1:44pm
Short Link:
Type:
Document Year:
Cite

Document Files

camready-1031.ppt

(213 downloads)

Keywords

Subscribe

[1] , "Novel data clustering for microarrays and image segmentation", IEEE SigPort, 2015. [Online]. Available: http://sigport.org/561. Accessed: Apr. 23, 2017.
@article{561-15,
url = {http://sigport.org/561},
author = { },
publisher = {IEEE SigPort},
title = {Novel data clustering for microarrays and image segmentation},
year = {2015} }
TY - EJOUR
T1 - Novel data clustering for microarrays and image segmentation
AU -
PY - 2015
PB - IEEE SigPort
UR - http://sigport.org/561
ER -
. (2015). Novel data clustering for microarrays and image segmentation. IEEE SigPort. http://sigport.org/561
, 2015. Novel data clustering for microarrays and image segmentation. Available at: http://sigport.org/561.
. (2015). "Novel data clustering for microarrays and image segmentation." Web.
1. . Novel data clustering for microarrays and image segmentation [Internet]. IEEE SigPort; 2015. Available from : http://sigport.org/561

Edge-enhancing filters with negative weights


Edge-enhanced eigenvectors of the Laplacian with a negative weight

In [doi{10.1109/ICMEW.2014.6890711}], a~graph-based filtering of noisy images is performed by directly computing a projection of the image to be filtered onto a lower dimensional Krylov subspace of the graph Laplacian, constructed using non-negative graph weights determined by distances between image data corresponding to image pixels. We extend the construction of the graph Laplacian to the case, where some graph weights can be negative.

KGlobalSIP.pdf

PDF icon KGlobalSIP.pdf (216 downloads)

Paper Details

Authors:
Submitted On:
23 February 2016 - 1:44pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

KGlobalSIP.pdf

(216 downloads)

Keywords

Subscribe

[1] , "Edge-enhancing filters with negative weights", IEEE SigPort, 2015. [Online]. Available: http://sigport.org/503. Accessed: Apr. 23, 2017.
@article{503-15,
url = {http://sigport.org/503},
author = { },
publisher = {IEEE SigPort},
title = {Edge-enhancing filters with negative weights},
year = {2015} }
TY - EJOUR
T1 - Edge-enhancing filters with negative weights
AU -
PY - 2015
PB - IEEE SigPort
UR - http://sigport.org/503
ER -
. (2015). Edge-enhancing filters with negative weights. IEEE SigPort. http://sigport.org/503
, 2015. Edge-enhancing filters with negative weights. Available at: http://sigport.org/503.
. (2015). "Edge-enhancing filters with negative weights." Web.
1. . Edge-enhancing filters with negative weights [Internet]. IEEE SigPort; 2015. Available from : http://sigport.org/503

Kernel-based low-rank feature extraction on a budget for Big data streams

Paper Details

Authors:
Dimitrios Berberidis, Georgios B. Giannakis
Submitted On:
23 February 2016 - 1:44pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

Globalsip2015.pdf

(427 downloads)

Keywords

Subscribe

[1] Dimitrios Berberidis, Georgios B. Giannakis, "Kernel-based low-rank feature extraction on a budget for Big data streams", IEEE SigPort, 2015. [Online]. Available: http://sigport.org/434. Accessed: Apr. 23, 2017.
@article{434-15,
url = {http://sigport.org/434},
author = {Dimitrios Berberidis; Georgios B. Giannakis },
publisher = {IEEE SigPort},
title = {Kernel-based low-rank feature extraction on a budget for Big data streams},
year = {2015} }
TY - EJOUR
T1 - Kernel-based low-rank feature extraction on a budget for Big data streams
AU - Dimitrios Berberidis; Georgios B. Giannakis
PY - 2015
PB - IEEE SigPort
UR - http://sigport.org/434
ER -
Dimitrios Berberidis, Georgios B. Giannakis. (2015). Kernel-based low-rank feature extraction on a budget for Big data streams. IEEE SigPort. http://sigport.org/434
Dimitrios Berberidis, Georgios B. Giannakis, 2015. Kernel-based low-rank feature extraction on a budget for Big data streams. Available at: http://sigport.org/434.
Dimitrios Berberidis, Georgios B. Giannakis. (2015). "Kernel-based low-rank feature extraction on a budget for Big data streams." Web.
1. Dimitrios Berberidis, Georgios B. Giannakis. Kernel-based low-rank feature extraction on a budget for Big data streams [Internet]. IEEE SigPort; 2015. Available from : http://sigport.org/434

Accelerated graph-based spectral polynomial filters


BF, GF and CG filters on 1D signals

Graph-based spectral denoising is a low-pass filtering using the eigendecomposition of the graph Laplacian matrix of a noisy signal. Polynomial filtering avoids costly computation of the eigendecomposition by projections onto suitable Krylov subspaces. Polynomial filters can be based, e.g., on the bilateral and guided filters. We propose constructing accelerated polynomial filters by running flexible Krylov subspace based linear and eigenvalue solvers such as the Block Locally Optimal Preconditioned Conjugate Gradient (LOBPCG) method.

MLSP2015.pdf

PDF icon MLSP2015.pdf (249 downloads)

Paper Details

Authors:
Alexander Malyshev
Submitted On:
23 February 2016 - 1:44pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

MLSP2015.pdf

(249 downloads)

Keywords

Subscribe

[1] Alexander Malyshev, "Accelerated graph-based spectral polynomial filters", IEEE SigPort, 2015. [Online]. Available: http://sigport.org/297. Accessed: Apr. 23, 2017.
@article{297-15,
url = {http://sigport.org/297},
author = {Alexander Malyshev },
publisher = {IEEE SigPort},
title = {Accelerated graph-based spectral polynomial filters},
year = {2015} }
TY - EJOUR
T1 - Accelerated graph-based spectral polynomial filters
AU - Alexander Malyshev
PY - 2015
PB - IEEE SigPort
UR - http://sigport.org/297
ER -
Alexander Malyshev. (2015). Accelerated graph-based spectral polynomial filters. IEEE SigPort. http://sigport.org/297
Alexander Malyshev, 2015. Accelerated graph-based spectral polynomial filters. Available at: http://sigport.org/297.
Alexander Malyshev. (2015). "Accelerated graph-based spectral polynomial filters." Web.
1. Alexander Malyshev. Accelerated graph-based spectral polynomial filters [Internet]. IEEE SigPort; 2015. Available from : http://sigport.org/297