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

Bayesian learning; Bayesian signal processing (MLR-BAYL)

EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS


Model confidence or uncertainty is critical in autonomous systems as they directly tie to the safety and trustworthiness of
the system. The quantification of uncertainty in the output decisions of deep neural networks (DNNs) is a challenging
problem. The Bayesian framework enables the estimation of the predictive uncertainty by introducing probability distributions
over the (unknown) network weights; however, the propagation of these high-dimensional distributions through

Paper Details

Authors:
Dimah Dera, Ghulam Rasool, and Nidhal Bouaynaya
Submitted On:
24 October 2019 - 8:37am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

IEEE MLSP presentation 2.pdf

(16)

Subscribe

[1] Dimah Dera, Ghulam Rasool, and Nidhal Bouaynaya, "EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4886. Accessed: Nov. 21, 2019.
@article{4886-19,
url = {http://sigport.org/4886},
author = {Dimah Dera; Ghulam Rasool; and Nidhal Bouaynaya },
publisher = {IEEE SigPort},
title = {EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS},
year = {2019} }
TY - EJOUR
T1 - EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS
AU - Dimah Dera; Ghulam Rasool; and Nidhal Bouaynaya
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4886
ER -
Dimah Dera, Ghulam Rasool, and Nidhal Bouaynaya. (2019). EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS. IEEE SigPort. http://sigport.org/4886
Dimah Dera, Ghulam Rasool, and Nidhal Bouaynaya, 2019. EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS. Available at: http://sigport.org/4886.
Dimah Dera, Ghulam Rasool, and Nidhal Bouaynaya. (2019). "EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS." Web.
1. Dimah Dera, Ghulam Rasool, and Nidhal Bouaynaya. EXTENDED VARIATIONAL INFERENCE FOR PROPAGATING UNCERTAINTY IN CONVOLUTIONAL NEURAL NETWORKS [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4886

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

(35)

Subscribe

[1] Meir Feder , "Minimax Active Learning via Minimal Model Capacity", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4876. Accessed: Nov. 21, 2019.
@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

Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method


In this article, we address the problem of estimating the state and learning of the parameters in a linear dynamic system with generalized $L_1$-regularization. Assuming a sparsity prior on the state, the joint state estimation and parameter learning problem is cast as an unconstrained optimization problem. However, when the dimensionality of state or parameters is large, memory requirements and computation of learning algorithms are generally prohibitive.

Paper Details

Authors:
Rui Gao, Filip Tronarp,Zheng Zhao, Simo Särkkä
Submitted On:
11 October 2019 - 11:29am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

mlsp_poster.pdf

(33)

Subscribe

[1] Rui Gao, Filip Tronarp,Zheng Zhao, Simo Särkkä , "Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4853. Accessed: Nov. 21, 2019.
@article{4853-19,
url = {http://sigport.org/4853},
author = {Rui Gao; Filip Tronarp;Zheng Zhao; Simo Särkkä },
publisher = {IEEE SigPort},
title = {Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method},
year = {2019} }
TY - EJOUR
T1 - Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method
AU - Rui Gao; Filip Tronarp;Zheng Zhao; Simo Särkkä
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4853
ER -
Rui Gao, Filip Tronarp,Zheng Zhao, Simo Särkkä . (2019). Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method. IEEE SigPort. http://sigport.org/4853
Rui Gao, Filip Tronarp,Zheng Zhao, Simo Särkkä , 2019. Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method. Available at: http://sigport.org/4853.
Rui Gao, Filip Tronarp,Zheng Zhao, Simo Särkkä . (2019). "Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method." Web.
1. Rui Gao, Filip Tronarp,Zheng Zhao, Simo Särkkä . Regularized state estimation and parameter learning via augmented Lagrangian Kalman smoother method [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4853

Sparse Bayesian Learning for Robust PCA

Paper Details

Authors:
Bhaskar Rao
Submitted On:
10 May 2019 - 5:46pm
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

SBL_RPCA.pdf

(38)

Subscribe

[1] Bhaskar Rao, "Sparse Bayesian Learning for Robust PCA", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4415. Accessed: Nov. 21, 2019.
@article{4415-19,
url = {http://sigport.org/4415},
author = {Bhaskar Rao },
publisher = {IEEE SigPort},
title = {Sparse Bayesian Learning for Robust PCA},
year = {2019} }
TY - EJOUR
T1 - Sparse Bayesian Learning for Robust PCA
AU - Bhaskar Rao
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4415
ER -
Bhaskar Rao. (2019). Sparse Bayesian Learning for Robust PCA. IEEE SigPort. http://sigport.org/4415
Bhaskar Rao, 2019. Sparse Bayesian Learning for Robust PCA. Available at: http://sigport.org/4415.
Bhaskar Rao. (2019). "Sparse Bayesian Learning for Robust PCA." Web.
1. Bhaskar Rao. Sparse Bayesian Learning for Robust PCA [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4415

Adversarial variational Bayes methods for Tweedie compound Poisson mixed models


The Tweedie Compound Poisson-Gamma model is routinely used for modeling non-negative continuous data with a discrete probability mass at zero. Mixed models with random effects account for the covariance structure related to the grouping hierarchy in the data. An important application of Tweedie mixed models is pricing the insurance policies, e.g. car insurance. However, the intractable likelihood function, the unknown variance function, and the hierarchical structure of mixed effects have presented considerable challenges for drawing inferences on Tweedie.

Paper Details

Authors:
Submitted On:
10 May 2019 - 1:24pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster.pdf

(39)

Subscribe

[1] , "Adversarial variational Bayes methods for Tweedie compound Poisson mixed models", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4375. Accessed: Nov. 21, 2019.
@article{4375-19,
url = {http://sigport.org/4375},
author = { },
publisher = {IEEE SigPort},
title = {Adversarial variational Bayes methods for Tweedie compound Poisson mixed models},
year = {2019} }
TY - EJOUR
T1 - Adversarial variational Bayes methods for Tweedie compound Poisson mixed models
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4375
ER -
. (2019). Adversarial variational Bayes methods for Tweedie compound Poisson mixed models. IEEE SigPort. http://sigport.org/4375
, 2019. Adversarial variational Bayes methods for Tweedie compound Poisson mixed models. Available at: http://sigport.org/4375.
. (2019). "Adversarial variational Bayes methods for Tweedie compound Poisson mixed models." Web.
1. . Adversarial variational Bayes methods for Tweedie compound Poisson mixed models [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4375

UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE

Paper Details

Authors:
Yanna Cavalcanti, Thomas Oberlin, Nicolas Dobigeon, Cédric Févotte, Simon Stute, Clovis Tauber
Submitted On:
10 May 2019 - 10:54am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster_icassp.pdf

(38)

Subscribe

[1] Yanna Cavalcanti, Thomas Oberlin, Nicolas Dobigeon, Cédric Févotte, Simon Stute, Clovis Tauber, "UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4350. Accessed: Nov. 21, 2019.
@article{4350-19,
url = {http://sigport.org/4350},
author = {Yanna Cavalcanti; Thomas Oberlin; Nicolas Dobigeon; Cédric Févotte; Simon Stute; Clovis Tauber },
publisher = {IEEE SigPort},
title = {UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE},
year = {2019} }
TY - EJOUR
T1 - UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE
AU - Yanna Cavalcanti; Thomas Oberlin; Nicolas Dobigeon; Cédric Févotte; Simon Stute; Clovis Tauber
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4350
ER -
Yanna Cavalcanti, Thomas Oberlin, Nicolas Dobigeon, Cédric Févotte, Simon Stute, Clovis Tauber. (2019). UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE. IEEE SigPort. http://sigport.org/4350
Yanna Cavalcanti, Thomas Oberlin, Nicolas Dobigeon, Cédric Févotte, Simon Stute, Clovis Tauber, 2019. UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE. Available at: http://sigport.org/4350.
Yanna Cavalcanti, Thomas Oberlin, Nicolas Dobigeon, Cédric Févotte, Simon Stute, Clovis Tauber. (2019). "UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE." Web.
1. Yanna Cavalcanti, Thomas Oberlin, Nicolas Dobigeon, Cédric Févotte, Simon Stute, Clovis Tauber. UNMIXING DYNAMIC PET IMAGES: COMBINING SPATIAL HETEROGENEITY AND NON-GAUSSIAN NOISE [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4350

Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method


Road traffic forecasting systems are in scenarios where sensor or system failure occur. In those scenarios, it is known that missing values negatively affect estimation accuracy although it is being often underestimate in current deep neural network approaches. Our assumption is that traffic data can be generated from a latent space. Thus, we propose an online unsupervised data imputation method based on learning the data distribution using a variational autoencoder (VAE).

Paper Details

Authors:
Guillem Boquet, Jose Lopez Vicario, Antoni Morell, Javier Serrano
Submitted On:
10 May 2019 - 7:11am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Slides_WIN_v2.pdf

(54)

Subscribe

[1] Guillem Boquet, Jose Lopez Vicario, Antoni Morell, Javier Serrano, "Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method ", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4303. Accessed: Nov. 21, 2019.
@article{4303-19,
url = {http://sigport.org/4303},
author = {Guillem Boquet; Jose Lopez Vicario; Antoni Morell; Javier Serrano },
publisher = {IEEE SigPort},
title = {Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method },
year = {2019} }
TY - EJOUR
T1 - Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method
AU - Guillem Boquet; Jose Lopez Vicario; Antoni Morell; Javier Serrano
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4303
ER -
Guillem Boquet, Jose Lopez Vicario, Antoni Morell, Javier Serrano. (2019). Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method . IEEE SigPort. http://sigport.org/4303
Guillem Boquet, Jose Lopez Vicario, Antoni Morell, Javier Serrano, 2019. Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method . Available at: http://sigport.org/4303.
Guillem Boquet, Jose Lopez Vicario, Antoni Morell, Javier Serrano. (2019). "Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method ." Web.
1. Guillem Boquet, Jose Lopez Vicario, Antoni Morell, Javier Serrano. Missing Data In Traffic Estimation: A Variational Autoencoder Imputation Method [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4303

ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models


Real-time nonlinear Bayesian filtering algorithms are overwhelmed by data volume, velocity and increasing complexity of computational models. In this paper, we propose a novel ensemble based nonlinear Bayesian filtering approach which only requires a small number of simulations and can be applied to high-dimensional systems in the presence of intractable likelihood functions.

Paper Details

Authors:
Submitted On:
8 May 2019 - 2:26pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models

(32)

Subscribe

[1] , "ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4128. Accessed: Nov. 21, 2019.
@article{4128-19,
url = {http://sigport.org/4128},
author = { },
publisher = {IEEE SigPort},
title = {ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models},
year = {2019} }
TY - EJOUR
T1 - ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4128
ER -
. (2019). ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models. IEEE SigPort. http://sigport.org/4128
, 2019. ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models. Available at: http://sigport.org/4128.
. (2019). "ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models." Web.
1. . ENLLVM: Ensemble based Nonlinear Bayesian Filtering using Linear Latent Variable Models [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4128

Variational and Hierarchical Recurrent Autoencoder


Despite a great success in learning representation for image data, it is challenging to learn the stochastic latent features from natural language based on variational inference. The difficulty in stochastic sequential learning is due to the posterior collapse caused by an autoregressive decoder which is prone to be too strong to learn sufficient latent information during optimization. To compensate this weakness in learning procedure, a sophisticated latent structure is required to assure good convergence so that random features are sufficiently captured for sequential decoding.

Paper Details

Authors:
Jen-Tzung Chien and Chun-Wei Wang
Submitted On:
7 May 2019 - 8:19pm
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

[ICASSP 2019] Variational and hierarchical recurrent autoencoder.pdf

(57)

Icassp19_hier.pdf

(36)

Subscribe

[1] Jen-Tzung Chien and Chun-Wei Wang, "Variational and Hierarchical Recurrent Autoencoder", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3968. Accessed: Nov. 21, 2019.
@article{3968-19,
url = {http://sigport.org/3968},
author = {Jen-Tzung Chien and Chun-Wei Wang },
publisher = {IEEE SigPort},
title = {Variational and Hierarchical Recurrent Autoencoder},
year = {2019} }
TY - EJOUR
T1 - Variational and Hierarchical Recurrent Autoencoder
AU - Jen-Tzung Chien and Chun-Wei Wang
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3968
ER -
Jen-Tzung Chien and Chun-Wei Wang. (2019). Variational and Hierarchical Recurrent Autoencoder. IEEE SigPort. http://sigport.org/3968
Jen-Tzung Chien and Chun-Wei Wang, 2019. Variational and Hierarchical Recurrent Autoencoder. Available at: http://sigport.org/3968.
Jen-Tzung Chien and Chun-Wei Wang. (2019). "Variational and Hierarchical Recurrent Autoencoder." Web.
1. Jen-Tzung Chien and Chun-Wei Wang. Variational and Hierarchical Recurrent Autoencoder [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3968

A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems


In spatial modulation (SM) systems that utilize multiple transmit antennas/patterns with a single radio front-end, we propose a learning approach to predict the average symbol error rate (SER) conditioned on the instantaneous channel state. We show that the predicted SER can be used to lower the average SER over Rayleigh fading channels by selecting the optimal codebook in each transmission instance.

Paper Details

Authors:
Baptiste Cavarec, Joakim Jaldén, Hugo Tullberg, Mats Bengtsson
Submitted On:
27 March 2019 - 9:05am
Short Link:
Type:
Document Year:
Cite

Document Files

Asilomar18_Poster_AdaptiveSM

(44)

Subscribe

[1] Baptiste Cavarec, Joakim Jaldén, Hugo Tullberg, Mats Bengtsson, "A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3669. Accessed: Nov. 21, 2019.
@article{3669-18,
url = {http://sigport.org/3669},
author = {Baptiste Cavarec; Joakim Jaldén; Hugo Tullberg; Mats Bengtsson },
publisher = {IEEE SigPort},
title = {A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems},
year = {2018} }
TY - EJOUR
T1 - A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems
AU - Baptiste Cavarec; Joakim Jaldén; Hugo Tullberg; Mats Bengtsson
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3669
ER -
Baptiste Cavarec, Joakim Jaldén, Hugo Tullberg, Mats Bengtsson. (2018). A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems. IEEE SigPort. http://sigport.org/3669
Baptiste Cavarec, Joakim Jaldén, Hugo Tullberg, Mats Bengtsson, 2018. A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems. Available at: http://sigport.org/3669.
Baptiste Cavarec, Joakim Jaldén, Hugo Tullberg, Mats Bengtsson. (2018). "A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems." Web.
1. Baptiste Cavarec, Joakim Jaldén, Hugo Tullberg, Mats Bengtsson. A Learning Approach for Optimal Codebook Selection in Spatial Modulation Systems [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3669

Pages