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

Neural network learning (MLR-NNLR)

ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION


Attention-based models have recently shown great performance on a range of tasks, such as speech recognition, machine translation, and image captioning due to their ability to summarize relevant information that expands through the entire length of an input sequence. In this paper, we analyze the usage of attention mechanisms to the problem of sequence summarization in our end-to-end text-dependent speaker recognition system. We explore different topologies and their variants of the attention layer, and compare different pooling methods on the attention weights.

Paper Details

Authors:
F A Rezaur Rahman Chowdhury, Quan Wang, Ignacio Lopez Moreno, Li Wan
Submitted On:
12 April 2018 - 11:42am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

icassp2018_poster_reza_attention

(26 downloads)

Keywords

Subscribe

[1] F A Rezaur Rahman Chowdhury, Quan Wang, Ignacio Lopez Moreno, Li Wan, "ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2413. Accessed: May. 20, 2018.
@article{2413-18,
url = {http://sigport.org/2413},
author = {F A Rezaur Rahman Chowdhury; Quan Wang; Ignacio Lopez Moreno; Li Wan },
publisher = {IEEE SigPort},
title = {ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION},
year = {2018} }
TY - EJOUR
T1 - ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION
AU - F A Rezaur Rahman Chowdhury; Quan Wang; Ignacio Lopez Moreno; Li Wan
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2413
ER -
F A Rezaur Rahman Chowdhury, Quan Wang, Ignacio Lopez Moreno, Li Wan. (2018). ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION. IEEE SigPort. http://sigport.org/2413
F A Rezaur Rahman Chowdhury, Quan Wang, Ignacio Lopez Moreno, Li Wan, 2018. ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION. Available at: http://sigport.org/2413.
F A Rezaur Rahman Chowdhury, Quan Wang, Ignacio Lopez Moreno, Li Wan. (2018). "ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION." Web.
1. F A Rezaur Rahman Chowdhury, Quan Wang, Ignacio Lopez Moreno, Li Wan. ATTENTION-BASED MODELS FOR TEXT-DEPENDENT SPEAKER VERIFICATION [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2413

Improving the Capacity of Very Deep Networks with Maxout Units


Deep neural networks inherently have large representational power for approximating complex target functions. However,

Paper Details

Authors:
Oyebade Oyedotun, Abd El Rahman Shabayek, Djamila Aouada, Björn Ottersten
Submitted On:
12 April 2018 - 11:43am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP_poster_Oyebade_V02.pdf

(12 downloads)

Keywords

Subscribe

[1] Oyebade Oyedotun, Abd El Rahman Shabayek, Djamila Aouada, Björn Ottersten, "Improving the Capacity of Very Deep Networks with Maxout Units", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2410. Accessed: May. 20, 2018.
@article{2410-18,
url = {http://sigport.org/2410},
author = {Oyebade Oyedotun; Abd El Rahman Shabayek; Djamila Aouada; Björn Ottersten },
publisher = {IEEE SigPort},
title = {Improving the Capacity of Very Deep Networks with Maxout Units},
year = {2018} }
TY - EJOUR
T1 - Improving the Capacity of Very Deep Networks with Maxout Units
AU - Oyebade Oyedotun; Abd El Rahman Shabayek; Djamila Aouada; Björn Ottersten
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2410
ER -
Oyebade Oyedotun, Abd El Rahman Shabayek, Djamila Aouada, Björn Ottersten. (2018). Improving the Capacity of Very Deep Networks with Maxout Units. IEEE SigPort. http://sigport.org/2410
Oyebade Oyedotun, Abd El Rahman Shabayek, Djamila Aouada, Björn Ottersten, 2018. Improving the Capacity of Very Deep Networks with Maxout Units. Available at: http://sigport.org/2410.
Oyebade Oyedotun, Abd El Rahman Shabayek, Djamila Aouada, Björn Ottersten. (2018). "Improving the Capacity of Very Deep Networks with Maxout Units." Web.
1. Oyebade Oyedotun, Abd El Rahman Shabayek, Djamila Aouada, Björn Ottersten. Improving the Capacity of Very Deep Networks with Maxout Units [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2410

HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION

Paper Details

Authors:
Submitted On:
12 April 2018 - 11:10am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

sampling_rates_poster.pdf

(19 downloads)

Keywords

Subscribe

[1] , "HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2380. Accessed: May. 20, 2018.
@article{2380-18,
url = {http://sigport.org/2380},
author = { },
publisher = {IEEE SigPort},
title = {HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION},
year = {2018} }
TY - EJOUR
T1 - HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION
AU -
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2380
ER -
. (2018). HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION. IEEE SigPort. http://sigport.org/2380
, 2018. HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION. Available at: http://sigport.org/2380.
. (2018). "HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION." Web.
1. . HOW SAMPLING RATE AFFECTS CROSS-DOMAIN TRANSFER LEARNING FOR VIDEO DESCRIPTION [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2380

AI: A Signal Processing Perspective


The signal processing (SP) landscape has been enriched by recent advances in artificial intelligence (AI) and machine learning (ML), especially since 2010 or so, yielding new tools for signal estimation, classification, prediction, and manipulation. Layered signal representations, nonlinear function approximation, and nonlinear signal prediction are now feasible at very large scale in both dimensionality and data size.

Paper Details

Authors:
Brian M. Sadler
Submitted On:
1 December 2017 - 9:19pm
Short Link:
Type:
Presenter's Name:
Document Year:
Cite

Document Files

Presentation slides (pdf version)

(132 downloads)

Keywords

Subscribe

[1] Brian M. Sadler, "AI: A Signal Processing Perspective", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2370. Accessed: May. 20, 2018.
@article{2370-17,
url = {http://sigport.org/2370},
author = {Brian M. Sadler },
publisher = {IEEE SigPort},
title = {AI: A Signal Processing Perspective},
year = {2017} }
TY - EJOUR
T1 - AI: A Signal Processing Perspective
AU - Brian M. Sadler
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2370
ER -
Brian M. Sadler. (2017). AI: A Signal Processing Perspective. IEEE SigPort. http://sigport.org/2370
Brian M. Sadler, 2017. AI: A Signal Processing Perspective. Available at: http://sigport.org/2370.
Brian M. Sadler. (2017). "AI: A Signal Processing Perspective." Web.
1. Brian M. Sadler. AI: A Signal Processing Perspective [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2370

Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs

Paper Details

Authors:
Jean-Charles Vialatte, Vincent Gripon, Gilles Coppin
Submitted On:
13 November 2017 - 12:29pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

slides.pdf

(377 downloads)

Keywords

Subscribe

[1] Jean-Charles Vialatte, Vincent Gripon, Gilles Coppin, "Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2339. Accessed: May. 20, 2018.
@article{2339-17,
url = {http://sigport.org/2339},
author = {Jean-Charles Vialatte; Vincent Gripon; Gilles Coppin },
publisher = {IEEE SigPort},
title = {Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs},
year = {2017} }
TY - EJOUR
T1 - Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs
AU - Jean-Charles Vialatte; Vincent Gripon; Gilles Coppin
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2339
ER -
Jean-Charles Vialatte, Vincent Gripon, Gilles Coppin. (2017). Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs. IEEE SigPort. http://sigport.org/2339
Jean-Charles Vialatte, Vincent Gripon, Gilles Coppin, 2017. Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs. Available at: http://sigport.org/2339.
Jean-Charles Vialatte, Vincent Gripon, Gilles Coppin. (2017). "Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs." Web.
1. Jean-Charles Vialatte, Vincent Gripon, Gilles Coppin. Learning Local Receptive Fields and their Weight Sharing Scheme on Graphs [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2339

ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION


Artery/vein classification in fundus images is a prerequisite for the assessment of diseases such as diabetes, hypertension or other cardiovascular pathologies. One clinical measure used to assess the severity of cardiovascular risk is the retinal arterio-venous ratio (AVR), which significantly depends on the accuracy of vessel classification into arteries or veins. This paper proposes a novel method for artery/vein classification combining deep learning and graph propagation strategies.

Paper Details

Authors:
Fantin Girard, Farida Cheriet
Submitted On:
11 November 2017 - 10:33am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:

Document Files

GlobalSIP 2017 slides

(90 downloads)

Keywords

Subscribe

[1] Fantin Girard, Farida Cheriet, "ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2307. Accessed: May. 20, 2018.
@article{2307-17,
url = {http://sigport.org/2307},
author = {Fantin Girard; Farida Cheriet },
publisher = {IEEE SigPort},
title = {ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION},
year = {2017} }
TY - EJOUR
T1 - ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION
AU - Fantin Girard; Farida Cheriet
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2307
ER -
Fantin Girard, Farida Cheriet. (2017). ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION. IEEE SigPort. http://sigport.org/2307
Fantin Girard, Farida Cheriet, 2017. ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION. Available at: http://sigport.org/2307.
Fantin Girard, Farida Cheriet. (2017). "ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION." Web.
1. Fantin Girard, Farida Cheriet. ARTERY/VEIN CLASSIFICATION IN FUNDUS IMAGES USING CNN AND LIKELIHOOD SCORE PROPAGATION [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2307

Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network


We present an efficient multi-image super resolution (MISR) method. Our solution consists of a L1-norm optimized reconstruction scheme for super resolution (SR), and a three-layer convolutional network for artifacts removal, in a concatenated fashion. Such a two-stage method achieves excellent performance, which outperforms the existing state-of-the-art SR methods in both subjective and objective measurements (e.g., 5 to 7 dB improvements on popular image database using PSNR metric).

Paper Details

Authors:
Jie Wu, Tao Yue, Qiu Shen, Xun Cao, Zhan Ma
Submitted On:
9 November 2017 - 10:11pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Super-resolution

(79 downloads)

Keywords

Subscribe

[1] Jie Wu, Tao Yue, Qiu Shen, Xun Cao, Zhan Ma, "Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2279. Accessed: May. 20, 2018.
@article{2279-17,
url = {http://sigport.org/2279},
author = {Jie Wu; Tao Yue; Qiu Shen; Xun Cao; Zhan Ma },
publisher = {IEEE SigPort},
title = {Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network},
year = {2017} }
TY - EJOUR
T1 - Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network
AU - Jie Wu; Tao Yue; Qiu Shen; Xun Cao; Zhan Ma
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2279
ER -
Jie Wu, Tao Yue, Qiu Shen, Xun Cao, Zhan Ma. (2017). Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network. IEEE SigPort. http://sigport.org/2279
Jie Wu, Tao Yue, Qiu Shen, Xun Cao, Zhan Ma, 2017. Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network. Available at: http://sigport.org/2279.
Jie Wu, Tao Yue, Qiu Shen, Xun Cao, Zhan Ma. (2017). "Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network." Web.
1. Jie Wu, Tao Yue, Qiu Shen, Xun Cao, Zhan Ma. Multiple-image Super Resolution Using Both Reconstruction Optimization and Deep Neural Network [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2279

When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks


Deep neural networks have led to dramatic improvements in performance for many machine learning tasks, yet the mathematical reasons for this success remain largely unclear. In this talk we present recent developments in the mathematical framework of convolutive neural networks (CNN). In particular we discuss the scattering network of Mallat and how it relates to another problem in harmonic analysis, namely the phase retrieval problem. Then we discuss the general convolutive neural network from a theoretician point of view.

Paper Details

Authors:
Radu Balan
Submitted On:
19 October 2017 - 11:56am
Short Link:
Type:
Document Year:
Cite

Document Files

Presentation slides (pdf version)

(150 downloads)

Keywords

Subscribe

[1] Radu Balan, "When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2263. Accessed: May. 20, 2018.
@article{2263-17,
url = {http://sigport.org/2263},
author = {Radu Balan },
publisher = {IEEE SigPort},
title = {When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks},
year = {2017} }
TY - EJOUR
T1 - When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks
AU - Radu Balan
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2263
ER -
Radu Balan. (2017). When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks. IEEE SigPort. http://sigport.org/2263
Radu Balan, 2017. When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks. Available at: http://sigport.org/2263.
Radu Balan. (2017). "When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks." Web.
1. Radu Balan. When Harmonic Analysis Meets Machine Learning: Lipschitz Analysis of Deep Convolution Networks [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2263

THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS

Paper Details

Authors:
Turgay Celik
Submitted On:
17 September 2017 - 9:18pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Klein-Poster.pdf

(166 downloads)

Keywords

Subscribe

[1] Turgay Celik, "THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2217. Accessed: May. 20, 2018.
@article{2217-17,
url = {http://sigport.org/2217},
author = {Turgay Celik },
publisher = {IEEE SigPort},
title = {THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS},
year = {2017} }
TY - EJOUR
T1 - THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS
AU - Turgay Celik
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2217
ER -
Turgay Celik. (2017). THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS. IEEE SigPort. http://sigport.org/2217
Turgay Celik, 2017. THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS. Available at: http://sigport.org/2217.
Turgay Celik. (2017). "THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS." Web.
1. Turgay Celik. THE WITS INTELLIGENT TEACHING SYSTEM: DETECTING STUDENT ENGAGEMENT DURING LECTURES USING CONVOLUTIONAL NEURAL NETWORKS [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2217

DenseNet for Dense Flow


Classical approaches for estimating optical flow have achieved rapid progress in the last decade. However, most of them are too slow to be applied in real-time video analysis. Due to the great success of deep learning, recent work has focused on using CNNs to solve such dense prediction problems. In this paper, we investigate a new deep architecture, Densely Connected Convolutional Networks (DenseNet), to learn optical flow. This specific architecture is ideal for the problem at hand as it provides shortcut connections throughout the network, which leads to implicit deep supervision.

Paper Details

Authors:
Yi Zhu,Shawn Newsam
Submitted On:
16 September 2017 - 2:45am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP17_paper2550_slides_yizhu.pdf

(195 downloads)

Keywords

Subscribe

[1] Yi Zhu,Shawn Newsam, "DenseNet for Dense Flow", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2181. Accessed: May. 20, 2018.
@article{2181-17,
url = {http://sigport.org/2181},
author = {Yi Zhu;Shawn Newsam },
publisher = {IEEE SigPort},
title = {DenseNet for Dense Flow},
year = {2017} }
TY - EJOUR
T1 - DenseNet for Dense Flow
AU - Yi Zhu;Shawn Newsam
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2181
ER -
Yi Zhu,Shawn Newsam. (2017). DenseNet for Dense Flow. IEEE SigPort. http://sigport.org/2181
Yi Zhu,Shawn Newsam, 2017. DenseNet for Dense Flow. Available at: http://sigport.org/2181.
Yi Zhu,Shawn Newsam. (2017). "DenseNet for Dense Flow." Web.
1. Yi Zhu,Shawn Newsam. DenseNet for Dense Flow [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2181

Pages