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

Learning theory and algorithms (MLR-LEAR)

Learn-by-Calibrating: Using Calibration as a Training Objective

Paper Details

Authors:
Jayaraman J. Thiagarajan, Bindya Venkatesh, Deepta Rajan
Submitted On:
26 May 2020 - 7:09pm
Short Link:
Type:

Document Files

lbc.pdf

(42)

Subscribe

[1] Jayaraman J. Thiagarajan, Bindya Venkatesh, Deepta Rajan, "Learn-by-Calibrating: Using Calibration as a Training Objective", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5439. Accessed: Oct. 24, 2020.
@article{5439-20,
url = {http://sigport.org/5439},
author = {Jayaraman J. Thiagarajan; Bindya Venkatesh; Deepta Rajan },
publisher = {IEEE SigPort},
title = {Learn-by-Calibrating: Using Calibration as a Training Objective},
year = {2020} }
TY - EJOUR
T1 - Learn-by-Calibrating: Using Calibration as a Training Objective
AU - Jayaraman J. Thiagarajan; Bindya Venkatesh; Deepta Rajan
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5439
ER -
Jayaraman J. Thiagarajan, Bindya Venkatesh, Deepta Rajan. (2020). Learn-by-Calibrating: Using Calibration as a Training Objective. IEEE SigPort. http://sigport.org/5439
Jayaraman J. Thiagarajan, Bindya Venkatesh, Deepta Rajan, 2020. Learn-by-Calibrating: Using Calibration as a Training Objective. Available at: http://sigport.org/5439.
Jayaraman J. Thiagarajan, Bindya Venkatesh, Deepta Rajan. (2020). "Learn-by-Calibrating: Using Calibration as a Training Objective." Web.
1. Jayaraman J. Thiagarajan, Bindya Venkatesh, Deepta Rajan. Learn-by-Calibrating: Using Calibration as a Training Objective [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5439

Learning Product Graphs from Multidomain Signals


In this paper, we focus on learning the underlying product graph structure from multidomain training data. We assume that the product graph is formed from a Cartesian graph product of two smaller factor graphs. We then pose the product graph learning problem as the factor graph Laplacian matrix estimation problem. To estimate the factor graph Laplacian matrices, we assume that the data is smooth with respect to the underlying product graph.

Paper Details

Authors:
Sai Kiran Kadambari, Sundeep Prabhakar Chepuri
Submitted On:
14 May 2020 - 7:38pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP20_Slides_sai.pdf

(49)

Subscribe

[1] Sai Kiran Kadambari, Sundeep Prabhakar Chepuri, "Learning Product Graphs from Multidomain Signals", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5325. Accessed: Oct. 24, 2020.
@article{5325-20,
url = {http://sigport.org/5325},
author = {Sai Kiran Kadambari; Sundeep Prabhakar Chepuri },
publisher = {IEEE SigPort},
title = {Learning Product Graphs from Multidomain Signals},
year = {2020} }
TY - EJOUR
T1 - Learning Product Graphs from Multidomain Signals
AU - Sai Kiran Kadambari; Sundeep Prabhakar Chepuri
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5325
ER -
Sai Kiran Kadambari, Sundeep Prabhakar Chepuri. (2020). Learning Product Graphs from Multidomain Signals. IEEE SigPort. http://sigport.org/5325
Sai Kiran Kadambari, Sundeep Prabhakar Chepuri, 2020. Learning Product Graphs from Multidomain Signals. Available at: http://sigport.org/5325.
Sai Kiran Kadambari, Sundeep Prabhakar Chepuri. (2020). "Learning Product Graphs from Multidomain Signals." Web.
1. Sai Kiran Kadambari, Sundeep Prabhakar Chepuri. Learning Product Graphs from Multidomain Signals [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5325

ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL


In this paper, we study the problem of online matrix completion (MC) aiming to achieve robustness to the variations in both low-rank subspace and noises. In contrast to existing methods, we progressively fit a specific Gaussian Mixture Model (GMM) for noises at each time slot, which ensures the adaptiveness of the model to dynamic complex noises under real application scenarios. Consequently, we formalize the online MC into an optimization problem based on the GMM regularizer.

Paper Details

Authors:
Chunsheng Liu, Chunlei Chen, Hong Shan, Bin Wang
Submitted On:
13 May 2020 - 11:11pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP.pdf

(37)

Subscribe

[1] Chunsheng Liu, Chunlei Chen, Hong Shan, Bin Wang, "ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5209. Accessed: Oct. 24, 2020.
@article{5209-20,
url = {http://sigport.org/5209},
author = {Chunsheng Liu; Chunlei Chen; Hong Shan; Bin Wang },
publisher = {IEEE SigPort},
title = {ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL},
year = {2020} }
TY - EJOUR
T1 - ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL
AU - Chunsheng Liu; Chunlei Chen; Hong Shan; Bin Wang
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5209
ER -
Chunsheng Liu, Chunlei Chen, Hong Shan, Bin Wang. (2020). ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL. IEEE SigPort. http://sigport.org/5209
Chunsheng Liu, Chunlei Chen, Hong Shan, Bin Wang, 2020. ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL. Available at: http://sigport.org/5209.
Chunsheng Liu, Chunlei Chen, Hong Shan, Bin Wang. (2020). "ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL." Web.
1. Chunsheng Liu, Chunlei Chen, Hong Shan, Bin Wang. ROBUST ONLINE MATRIX COMPLETION WITH GAUSSIAN MIXTURE MODEL [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5209

Self-supervised Learning for ECG-based Emotion Recognition


We present an electrocardiogram (ECG) -based emotion recognition system using self-supervised learning. Our proposed architecture consists of two main networks, a signal transformation recognition network and an emotion recognition network. First, unlabelled data are used to successfully train the former network to detect specific pre-determined signal transformations in the self-supervised learning step.

Paper Details

Authors:
Ali Etemad
Submitted On:
13 May 2020 - 6:57pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

icassp_sarkar2019self.pdf

(41)

Subscribe

[1] Ali Etemad, "Self-supervised Learning for ECG-based Emotion Recognition", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5165. Accessed: Oct. 24, 2020.
@article{5165-20,
url = {http://sigport.org/5165},
author = {Ali Etemad },
publisher = {IEEE SigPort},
title = {Self-supervised Learning for ECG-based Emotion Recognition},
year = {2020} }
TY - EJOUR
T1 - Self-supervised Learning for ECG-based Emotion Recognition
AU - Ali Etemad
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5165
ER -
Ali Etemad. (2020). Self-supervised Learning for ECG-based Emotion Recognition. IEEE SigPort. http://sigport.org/5165
Ali Etemad, 2020. Self-supervised Learning for ECG-based Emotion Recognition. Available at: http://sigport.org/5165.
Ali Etemad. (2020). "Self-supervised Learning for ECG-based Emotion Recognition." Web.
1. Ali Etemad. Self-supervised Learning for ECG-based Emotion Recognition [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5165

ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'


Tensor decompositions have become a central tool in machine learning to extract interpretable patterns from multiway arrays of data. However, computing the approximate Canonical Polyadic Decomposition (aCPD), one of the most important tensor decomposition model, remains a challenge. In this work, we propose several algorithms based on extrapolation that improve over existing alternating methods for aCPD.

Paper Details

Authors:
Andersen M. S. Ang, Jérémy Emile Cohen, Le Thi Khanh Hien, Nicolas Gillis
Submitted On:
13 May 2020 - 5:42pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Slide of ICASSP2020 presentation

(52)

Subscribe

[1] Andersen M. S. Ang, Jérémy Emile Cohen, Le Thi Khanh Hien, Nicolas Gillis, "ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5155. Accessed: Oct. 24, 2020.
@article{5155-20,
url = {http://sigport.org/5155},
author = {Andersen M. S. Ang; Jérémy Emile Cohen; Le Thi Khanh Hien; Nicolas Gillis },
publisher = {IEEE SigPort},
title = {ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'},
year = {2020} }
TY - EJOUR
T1 - ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'
AU - Andersen M. S. Ang; Jérémy Emile Cohen; Le Thi Khanh Hien; Nicolas Gillis
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5155
ER -
Andersen M. S. Ang, Jérémy Emile Cohen, Le Thi Khanh Hien, Nicolas Gillis. (2020). ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'. IEEE SigPort. http://sigport.org/5155
Andersen M. S. Ang, Jérémy Emile Cohen, Le Thi Khanh Hien, Nicolas Gillis, 2020. ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'. Available at: http://sigport.org/5155.
Andersen M. S. Ang, Jérémy Emile Cohen, Le Thi Khanh Hien, Nicolas Gillis. (2020). "ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION'." Web.
1. Andersen M. S. Ang, Jérémy Emile Cohen, Le Thi Khanh Hien, Nicolas Gillis. ICASSP 2020 presentation slide of 'EXTRAPOLATED ALTERNATING ALGORITHMS FOR APPROXIMATE CANONICAL POLYADIC DECOMPOSITION' [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5155

Generalized Kernel-Based Dynamic Mode Decomposition

Paper Details

Authors:
Patrick Héas, Cédric Herzet, Benoit Combès
Submitted On:
11 February 2020 - 8:21am
Short Link:
Type:
Event:

Document Files

manuscript.pdf

(75)

Subscribe

[1] Patrick Héas, Cédric Herzet, Benoit Combès, "Generalized Kernel-Based Dynamic Mode Decomposition", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/4978. Accessed: Oct. 24, 2020.
@article{4978-20,
url = {http://sigport.org/4978},
author = {Patrick Héas; Cédric Herzet; Benoit Combès },
publisher = {IEEE SigPort},
title = {Generalized Kernel-Based Dynamic Mode Decomposition},
year = {2020} }
TY - EJOUR
T1 - Generalized Kernel-Based Dynamic Mode Decomposition
AU - Patrick Héas; Cédric Herzet; Benoit Combès
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/4978
ER -
Patrick Héas, Cédric Herzet, Benoit Combès. (2020). Generalized Kernel-Based Dynamic Mode Decomposition. IEEE SigPort. http://sigport.org/4978
Patrick Héas, Cédric Herzet, Benoit Combès, 2020. Generalized Kernel-Based Dynamic Mode Decomposition. Available at: http://sigport.org/4978.
Patrick Héas, Cédric Herzet, Benoit Combès. (2020). "Generalized Kernel-Based Dynamic Mode Decomposition." Web.
1. Patrick Héas, Cédric Herzet, Benoit Combès. Generalized Kernel-Based Dynamic Mode Decomposition [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/4978

Serious Games and ML for Detecting MCI


Our work has focused on detecting Mild Cognitive Impairment (MCI) by developing Serious Games (SG) on mobile devices, distinct from games marketed as 'brain training' which claim to maintain mental acuity. One game, WarCAT, captures players' moves during the game to infer processes of strategy recognition, learning, and memory. The purpose of our game is to use the generated game-play data combined with machine learning (ML) to help detect MCI. MCI is difficult to detect for several reasons.

Paper Details

Authors:
Mahmood Aljumaili, Robert D McLeod, Marcia Friesen
Submitted On:
12 November 2019 - 12:45am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

GlobalSIP 2019 - Serious Games and ML for Detecting MCI.pdf

(153)

Subscribe

[1] Mahmood Aljumaili, Robert D McLeod, Marcia Friesen, "Serious Games and ML for Detecting MCI", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4949. Accessed: Oct. 24, 2020.
@article{4949-19,
url = {http://sigport.org/4949},
author = {Mahmood Aljumaili; Robert D McLeod; Marcia Friesen },
publisher = {IEEE SigPort},
title = {Serious Games and ML for Detecting MCI},
year = {2019} }
TY - EJOUR
T1 - Serious Games and ML for Detecting MCI
AU - Mahmood Aljumaili; Robert D McLeod; Marcia Friesen
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4949
ER -
Mahmood Aljumaili, Robert D McLeod, Marcia Friesen. (2019). Serious Games and ML for Detecting MCI. IEEE SigPort. http://sigport.org/4949
Mahmood Aljumaili, Robert D McLeod, Marcia Friesen, 2019. Serious Games and ML for Detecting MCI. Available at: http://sigport.org/4949.
Mahmood Aljumaili, Robert D McLeod, Marcia Friesen. (2019). "Serious Games and ML for Detecting MCI." Web.
1. Mahmood Aljumaili, Robert D McLeod, Marcia Friesen. Serious Games and ML for Detecting MCI [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4949

Wave Physics Informed Dictionary Learning in One Dimension

Paper Details

Authors:
Harsha Vardhan Tetali, K. Supreet Alguri, Joel B. Harley
Submitted On:
25 October 2019 - 1:50pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:

Document Files

HarshaMLSP.pdf

(120)

Subscribe

[1] Harsha Vardhan Tetali, K. Supreet Alguri, Joel B. Harley, "Wave Physics Informed Dictionary Learning in One Dimension", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4895. Accessed: Oct. 24, 2020.
@article{4895-19,
url = {http://sigport.org/4895},
author = {Harsha Vardhan Tetali; K. Supreet Alguri; Joel B. Harley },
publisher = {IEEE SigPort},
title = {Wave Physics Informed Dictionary Learning in One Dimension},
year = {2019} }
TY - EJOUR
T1 - Wave Physics Informed Dictionary Learning in One Dimension
AU - Harsha Vardhan Tetali; K. Supreet Alguri; Joel B. Harley
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4895
ER -
Harsha Vardhan Tetali, K. Supreet Alguri, Joel B. Harley. (2019). Wave Physics Informed Dictionary Learning in One Dimension. IEEE SigPort. http://sigport.org/4895
Harsha Vardhan Tetali, K. Supreet Alguri, Joel B. Harley, 2019. Wave Physics Informed Dictionary Learning in One Dimension. Available at: http://sigport.org/4895.
Harsha Vardhan Tetali, K. Supreet Alguri, Joel B. Harley. (2019). "Wave Physics Informed Dictionary Learning in One Dimension." Web.
1. Harsha Vardhan Tetali, K. Supreet Alguri, Joel B. Harley. Wave Physics Informed Dictionary Learning in One Dimension [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4895

Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization)


In this paper, we derive generic bounds on the maximum deviations in prediction errors for sequential prediction via an information-theoretic approach. The fundamental bounds are shown to depend only on the conditional entropy of the data point to be predicted given the previous data points. In the asymptotic case, the bounds are achieved if and only if the prediction error is white and uniformly distributed.

Paper Details

Authors:
Song Fang, Quanyan Zhu
Submitted On:
24 October 2019 - 4:45pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

postertemplate.pdf

(122)

Subscribe

[1] Song Fang, Quanyan Zhu, "Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization)", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4890. Accessed: Oct. 24, 2020.
@article{4890-19,
url = {http://sigport.org/4890},
author = {Song Fang; Quanyan Zhu },
publisher = {IEEE SigPort},
title = {Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization)},
year = {2019} }
TY - EJOUR
T1 - Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization)
AU - Song Fang; Quanyan Zhu
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4890
ER -
Song Fang, Quanyan Zhu. (2019). Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization). IEEE SigPort. http://sigport.org/4890
Song Fang, Quanyan Zhu, 2019. Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization). Available at: http://sigport.org/4890.
Song Fang, Quanyan Zhu. (2019). "Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization)." Web.
1. Song Fang, Quanyan Zhu. Generic Bounds on the Maximum Deviations in Sequential/Sequence Prediction (and the Implications in Recursive Algorithms and Learning/Generalization) [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4890

Minimax Active Learning via Minimal Model Capacity


Active learning is a form of machine learning which combines supervised learning and feedback to minimize the training set size, subject to low generalization errors. Since direct optimization of the generalization error is difficult, many heuristics have been developed which lack a firm theoretical foundation. In this paper, a new information theoretic criterion is proposed based on a minimax log-loss regret formulation of the active learning problem. In the first part of this paper, a Redundancy Capacity theorem for active learning is derived along with an optimal learner.

Paper Details

Authors:
Meir Feder
Submitted On:
16 October 2019 - 4:02pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

MLSP_2019_Minimax_Active_Learning_via_Minimal_Model_Capacity.pdf

(143)

Subscribe

[1] Meir Feder , "Minimax Active Learning via Minimal Model Capacity", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4876. Accessed: Oct. 24, 2020.
@article{4876-19,
url = {http://sigport.org/4876},
author = {Meir Feder },
publisher = {IEEE SigPort},
title = {Minimax Active Learning via Minimal Model Capacity},
year = {2019} }
TY - EJOUR
T1 - Minimax Active Learning via Minimal Model Capacity
AU - Meir Feder
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4876
ER -
Meir Feder . (2019). Minimax Active Learning via Minimal Model Capacity. IEEE SigPort. http://sigport.org/4876
Meir Feder , 2019. Minimax Active Learning via Minimal Model Capacity. Available at: http://sigport.org/4876.
Meir Feder . (2019). "Minimax Active Learning via Minimal Model Capacity." Web.
1. Meir Feder . Minimax Active Learning via Minimal Model Capacity [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4876

Pages