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

Learning theory and algorithms (MLR-LEAR)

Statistical rank selection for incomplete low-rank matrices

Paper Details

Authors:
Rui Zhang, Alexander Shapiro, Yao Xie
Submitted On:
15 May 2019 - 7:09pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:

Document Files

ICASSP2019.pdf

(4)

Subscribe

[1] Rui Zhang, Alexander Shapiro, Yao Xie, "Statistical rank selection for incomplete low-rank matrices", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4534. Accessed: May. 26, 2019.
@article{4534-19,
url = {http://sigport.org/4534},
author = {Rui Zhang; Alexander Shapiro; Yao Xie },
publisher = {IEEE SigPort},
title = {Statistical rank selection for incomplete low-rank matrices},
year = {2019} }
TY - EJOUR
T1 - Statistical rank selection for incomplete low-rank matrices
AU - Rui Zhang; Alexander Shapiro; Yao Xie
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4534
ER -
Rui Zhang, Alexander Shapiro, Yao Xie. (2019). Statistical rank selection for incomplete low-rank matrices. IEEE SigPort. http://sigport.org/4534
Rui Zhang, Alexander Shapiro, Yao Xie, 2019. Statistical rank selection for incomplete low-rank matrices. Available at: http://sigport.org/4534.
Rui Zhang, Alexander Shapiro, Yao Xie. (2019). "Statistical rank selection for incomplete low-rank matrices." Web.
1. Rui Zhang, Alexander Shapiro, Yao Xie. Statistical rank selection for incomplete low-rank matrices [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4534

Statistical rank selection for incomplete low-rank matrices

Paper Details

Authors:
Rui Zhang, Alexander Shapiro, Yao Xie
Submitted On:
15 May 2019 - 7:09pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:

Document Files

ICASSP2019.pdf

(4)

Subscribe

[1] Rui Zhang, Alexander Shapiro, Yao Xie, "Statistical rank selection for incomplete low-rank matrices", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4533. Accessed: May. 26, 2019.
@article{4533-19,
url = {http://sigport.org/4533},
author = {Rui Zhang; Alexander Shapiro; Yao Xie },
publisher = {IEEE SigPort},
title = {Statistical rank selection for incomplete low-rank matrices},
year = {2019} }
TY - EJOUR
T1 - Statistical rank selection for incomplete low-rank matrices
AU - Rui Zhang; Alexander Shapiro; Yao Xie
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4533
ER -
Rui Zhang, Alexander Shapiro, Yao Xie. (2019). Statistical rank selection for incomplete low-rank matrices. IEEE SigPort. http://sigport.org/4533
Rui Zhang, Alexander Shapiro, Yao Xie, 2019. Statistical rank selection for incomplete low-rank matrices. Available at: http://sigport.org/4533.
Rui Zhang, Alexander Shapiro, Yao Xie. (2019). "Statistical rank selection for incomplete low-rank matrices." Web.
1. Rui Zhang, Alexander Shapiro, Yao Xie. Statistical rank selection for incomplete low-rank matrices [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4533

A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties


Stochastic mirror descent (SMD) algorithms have recently garnered a great deal of attention in optimization, signal processing, and machine learning. They are similar to stochastic gradient descent (SGD), in that they perform updates along the negative gradient of an instantaneous (or stochastically chosen) loss function. However, rather than update the parameter (or weight) vector directly, they update it in a "mirrored" domain whose transformation is given by the gradient of a strictly convex differentiable potential function.

Paper Details

Authors:
Navid Azizan, Babak Hassibi
Submitted On:
13 May 2019 - 8:33pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP-SMD-Poster.pdf

(8)

Subscribe

[1] Navid Azizan, Babak Hassibi, "A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4498. Accessed: May. 26, 2019.
@article{4498-19,
url = {http://sigport.org/4498},
author = {Navid Azizan; Babak Hassibi },
publisher = {IEEE SigPort},
title = {A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties},
year = {2019} }
TY - EJOUR
T1 - A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties
AU - Navid Azizan; Babak Hassibi
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4498
ER -
Navid Azizan, Babak Hassibi. (2019). A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties. IEEE SigPort. http://sigport.org/4498
Navid Azizan, Babak Hassibi, 2019. A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties. Available at: http://sigport.org/4498.
Navid Azizan, Babak Hassibi. (2019). "A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties." Web.
1. Navid Azizan, Babak Hassibi. A Characterization of Stochastic Mirror Descent Algorithms and Their Convergence Properties [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4498

A Fast Method of Computing Persistent Homology of Time Series Data

Paper Details

Authors:
Kazuyuki Aihara
Submitted On:
10 May 2019 - 10:46am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP2019_poster_tsuji_20190508.pdf

(1)

Subscribe

[1] Kazuyuki Aihara, "A Fast Method of Computing Persistent Homology of Time Series Data", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4348. Accessed: May. 26, 2019.
@article{4348-19,
url = {http://sigport.org/4348},
author = {Kazuyuki Aihara },
publisher = {IEEE SigPort},
title = {A Fast Method of Computing Persistent Homology of Time Series Data},
year = {2019} }
TY - EJOUR
T1 - A Fast Method of Computing Persistent Homology of Time Series Data
AU - Kazuyuki Aihara
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4348
ER -
Kazuyuki Aihara. (2019). A Fast Method of Computing Persistent Homology of Time Series Data. IEEE SigPort. http://sigport.org/4348
Kazuyuki Aihara, 2019. A Fast Method of Computing Persistent Homology of Time Series Data. Available at: http://sigport.org/4348.
Kazuyuki Aihara. (2019). "A Fast Method of Computing Persistent Homology of Time Series Data." Web.
1. Kazuyuki Aihara. A Fast Method of Computing Persistent Homology of Time Series Data [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4348

Pairwise Approximate K-SVD


Pairwise, or separable, dictionaries are suited for the sparse representation of 2D signals in their original form, without vectorization. They are equivalent with enforcing a Kronecker structure on a standard dictionary for 1D signals. We present a dictionary learning algorithm, in the coordinate descent style of Approximate K-SVD, for such dictionaries. The algorithm has the benefit of extremely low complexity, clearly lower than that of existing algorithms.

Paper Details

Authors:
Paul Irofti, Bogdan Dumitrescu
Submitted On:
10 May 2019 - 3:58am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

pair-ksvd-poster.pdf

(3)

Subscribe

[1] Paul Irofti, Bogdan Dumitrescu, "Pairwise Approximate K-SVD", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4284. Accessed: May. 26, 2019.
@article{4284-19,
url = {http://sigport.org/4284},
author = {Paul Irofti; Bogdan Dumitrescu },
publisher = {IEEE SigPort},
title = {Pairwise Approximate K-SVD},
year = {2019} }
TY - EJOUR
T1 - Pairwise Approximate K-SVD
AU - Paul Irofti; Bogdan Dumitrescu
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4284
ER -
Paul Irofti, Bogdan Dumitrescu. (2019). Pairwise Approximate K-SVD. IEEE SigPort. http://sigport.org/4284
Paul Irofti, Bogdan Dumitrescu, 2019. Pairwise Approximate K-SVD. Available at: http://sigport.org/4284.
Paul Irofti, Bogdan Dumitrescu. (2019). "Pairwise Approximate K-SVD." Web.
1. Paul Irofti, Bogdan Dumitrescu. Pairwise Approximate K-SVD [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4284

Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information


Semidefinite programming has been shown to be both efficient and asymptotically optimal in solving community detection problems, as long as observations are purely graphical in nature. In this paper, we extend this result to observations that have both a graphical and a non-graphical component. We consider the binary censored block model with $n$ nodes and study the effect of partially revealed labels on the performance of semidefinite programming.

Paper Details

Authors:
Mohammad Esmaeili, Hussein Saad, Aria Nosratinia
Submitted On:
9 May 2019 - 9:07pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Esmaeili_ICASSP_Poster.pdf

(3)

Subscribe

[1] Mohammad Esmaeili, Hussein Saad, Aria Nosratinia, "Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4248. Accessed: May. 26, 2019.
@article{4248-19,
url = {http://sigport.org/4248},
author = {Mohammad Esmaeili; Hussein Saad; Aria Nosratinia },
publisher = {IEEE SigPort},
title = {Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information},
year = {2019} }
TY - EJOUR
T1 - Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information
AU - Mohammad Esmaeili; Hussein Saad; Aria Nosratinia
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4248
ER -
Mohammad Esmaeili, Hussein Saad, Aria Nosratinia. (2019). Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information. IEEE SigPort. http://sigport.org/4248
Mohammad Esmaeili, Hussein Saad, Aria Nosratinia, 2019. Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information. Available at: http://sigport.org/4248.
Mohammad Esmaeili, Hussein Saad, Aria Nosratinia. (2019). "Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information." Web.
1. Mohammad Esmaeili, Hussein Saad, Aria Nosratinia. Exact Recovery by Semidefinite Programming in the Binary Stochastic Block Model with Partially Revealed Side Information [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4248

Feature Selection for Multi-labeled Variables via Dependency Maximization


Feature selection and reducing the dimensionality of data is an essential step in data analysis. In this work, we propose a new criterion for feature selection that is formulated as conditional information between features given the labeled variable. Instead of using the standard mutual information measure based on Kullback-Leibler divergence, we use our proposed criterion to filter out redundant features for the purpose of multiclass classification.

Paper Details

Authors:
Salimeh Yasaei Sekeh, Alfred O. Hero
Submitted On:
9 May 2019 - 9:36am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP2019-Salimeh-V2.pdf

(3)

Subscribe

[1] Salimeh Yasaei Sekeh, Alfred O. Hero, "Feature Selection for Multi-labeled Variables via Dependency Maximization", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4202. Accessed: May. 26, 2019.
@article{4202-19,
url = {http://sigport.org/4202},
author = {Salimeh Yasaei Sekeh; Alfred O. Hero },
publisher = {IEEE SigPort},
title = {Feature Selection for Multi-labeled Variables via Dependency Maximization},
year = {2019} }
TY - EJOUR
T1 - Feature Selection for Multi-labeled Variables via Dependency Maximization
AU - Salimeh Yasaei Sekeh; Alfred O. Hero
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4202
ER -
Salimeh Yasaei Sekeh, Alfred O. Hero. (2019). Feature Selection for Multi-labeled Variables via Dependency Maximization. IEEE SigPort. http://sigport.org/4202
Salimeh Yasaei Sekeh, Alfred O. Hero, 2019. Feature Selection for Multi-labeled Variables via Dependency Maximization. Available at: http://sigport.org/4202.
Salimeh Yasaei Sekeh, Alfred O. Hero. (2019). "Feature Selection for Multi-labeled Variables via Dependency Maximization." Web.
1. Salimeh Yasaei Sekeh, Alfred O. Hero. Feature Selection for Multi-labeled Variables via Dependency Maximization [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4202

PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES


Clustering and categorization of similar images using SIFT and SURF require a high computational cost. In this paper, a simple approach to reduce the cardinality of keypoint set and prune the dimension of SIFT and SURF feature descriptors for efficient image clustering is proposed. For this purpose, sparsely spaced (uniformly distributed) important keypoints are chosen. In addition, multiple reduced dimensional variants of SIFT and SURF descriptors are presented.

Paper Details

Authors:
Tushar Shankar Shinde, Anil Kumar Tiwari
Submitted On:
9 May 2019 - 12:11am
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

ICASSP_Poster.pdf

(656)

Subscribe

[1] Tushar Shankar Shinde, Anil Kumar Tiwari, "PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4145. Accessed: May. 26, 2019.
@article{4145-19,
url = {http://sigport.org/4145},
author = {Tushar Shankar Shinde; Anil Kumar Tiwari },
publisher = {IEEE SigPort},
title = {PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES},
year = {2019} }
TY - EJOUR
T1 - PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES
AU - Tushar Shankar Shinde; Anil Kumar Tiwari
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4145
ER -
Tushar Shankar Shinde, Anil Kumar Tiwari. (2019). PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES. IEEE SigPort. http://sigport.org/4145
Tushar Shankar Shinde, Anil Kumar Tiwari, 2019. PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES. Available at: http://sigport.org/4145.
Tushar Shankar Shinde, Anil Kumar Tiwari. (2019). "PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES." Web.
1. Tushar Shankar Shinde, Anil Kumar Tiwari. PRUNING SIFT & SURF FOR EFFICIENT CLUSTERING OF NEAR-DUPLICATE IMAGES [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4145

Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices


We tackle the problem of recovering a complex signal $\mathbf{x}\in\mathbb{C}^n$ from quadratic measurements of the form $y_i=\mathbf{x}^*\mathbf{A}_i\mathbf{x}$, where $\{\mathbf{A}_i\}_{i=1}^m$ is a set of complex iid standard Gaussian matrices. This non-convex problem is related to the well understood phase retrieval problem where $\mathbf{A}_i$ is a rank-1 positive semidefinite matrix.

Paper Details

Authors:
Sidharth Gupta, Ivan Dokmanić
Submitted On:
8 May 2019 - 3:11pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

quadeq_icassp.pdf

(5)

Subscribe

[1] Sidharth Gupta, Ivan Dokmanić, "Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4132. Accessed: May. 26, 2019.
@article{4132-19,
url = {http://sigport.org/4132},
author = {Sidharth Gupta; Ivan Dokmanić },
publisher = {IEEE SigPort},
title = {Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices},
year = {2019} }
TY - EJOUR
T1 - Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices
AU - Sidharth Gupta; Ivan Dokmanić
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4132
ER -
Sidharth Gupta, Ivan Dokmanić. (2019). Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices. IEEE SigPort. http://sigport.org/4132
Sidharth Gupta, Ivan Dokmanić, 2019. Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices. Available at: http://sigport.org/4132.
Sidharth Gupta, Ivan Dokmanić. (2019). "Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices." Web.
1. Sidharth Gupta, Ivan Dokmanić. Solving Complex Quadratic Equations with Full-rank Random Gaussian Matrices [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4132

TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION


Kernel-based adaptive filters are sequential learning algorithms, operating on reproducing kernel Hilbert spaces. Their learning performance is susceptible to the selection of appropriate values for kernel bandwidth and learning-rate parameters. Additionally, as these algorithms train the model using a sequence of input vectors, their computation scales with the number of samples. We propose a framework that addresses the previous open challenges of kernel-based adaptive filters.

Paper Details

Authors:
Submitted On:
8 May 2019 - 3:01pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster_ICASSP2019.pdf

(7)

Subscribe

[1] , "TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4131. Accessed: May. 26, 2019.
@article{4131-19,
url = {http://sigport.org/4131},
author = { },
publisher = {IEEE SigPort},
title = {TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION},
year = {2019} }
TY - EJOUR
T1 - TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4131
ER -
. (2019). TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION. IEEE SigPort. http://sigport.org/4131
, 2019. TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION. Available at: http://sigport.org/4131.
. (2019). "TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION." Web.
1. . TIME SERIES PREDICTION FOR KERNEL-BASED ADAPTIVE FILTERS USING VARIABLE BANDWIDTH, ADAPTIVE LEARNING-RATE, AND DIMENSIONALITY REDUCTION [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4131

Pages