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

Machine Learning for Signal Processing

NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION


Obtaining aligned spectral pairs in case of non-parallel data for stand-alone Voice Conversion (VC) technique is a challenging research problem. Unsupervised alignment algorithm, namely, an Iterative combination of a Nearest Neighbor search step and a Conversion step Alignment (INCA) iteratively tries to align the spectral features by minimizing the Euclidean distance metric between the intermediate converted and the target spectral feature vectors.

Paper Details

Authors:
Nirmesh Shah, Hemant A. Patil
Submitted On:
8 May 2019 - 8:07am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

main.pdf

(23)

Subscribe

[1] Nirmesh Shah, Hemant A. Patil, "NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4080. Accessed: Oct. 16, 2019.
@article{4080-19,
url = {http://sigport.org/4080},
author = {Nirmesh Shah; Hemant A. Patil },
publisher = {IEEE SigPort},
title = {NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION},
year = {2019} }
TY - EJOUR
T1 - NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION
AU - Nirmesh Shah; Hemant A. Patil
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4080
ER -
Nirmesh Shah, Hemant A. Patil. (2019). NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION. IEEE SigPort. http://sigport.org/4080
Nirmesh Shah, Hemant A. Patil, 2019. NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION. Available at: http://sigport.org/4080.
Nirmesh Shah, Hemant A. Patil. (2019). "NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION." Web.
1. Nirmesh Shah, Hemant A. Patil. NOVEL METRIC LEARNING FOR NON-PARALLEL VOICE CONVERSION [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4080

Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition


Real-world data exhibiting high order/dimensionality and various couplings are linked to each other since they share
some common characteristics. Coupled tensor decomposition has become a popular technique for group analysis in recent
years, especially for simultaneous analysis of multi-block tensor data with common information. To address the multi-
block tensor data, we propose a fast double-coupled nonnegative Canonical Polyadic Decomposition (FDC-NCPD)

Paper Details

Authors:
Tapani Ristaniemi, Fengyu Cong
Submitted On:
8 May 2019 - 6:25am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster_Xiulin Wang.pdf

(26)

Keywords

Additional Categories

Subscribe

[1] Tapani Ristaniemi, Fengyu Cong, "Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4067. Accessed: Oct. 16, 2019.
@article{4067-19,
url = {http://sigport.org/4067},
author = {Tapani Ristaniemi; Fengyu Cong },
publisher = {IEEE SigPort},
title = {Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition},
year = {2019} }
TY - EJOUR
T1 - Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition
AU - Tapani Ristaniemi; Fengyu Cong
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4067
ER -
Tapani Ristaniemi, Fengyu Cong. (2019). Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition. IEEE SigPort. http://sigport.org/4067
Tapani Ristaniemi, Fengyu Cong, 2019. Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition. Available at: http://sigport.org/4067.
Tapani Ristaniemi, Fengyu Cong. (2019). "Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition." Web.
1. Tapani Ristaniemi, Fengyu Cong. Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4067

Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition


Real-world data exhibiting high order/dimensionality and various couplings are linked to each other since they share
some common characteristics. Coupled tensor decomposition has become a popular technique for group analysis in recent
years, especially for simultaneous analysis of multi-block tensor data with common information. To address the multi-
block tensor data, we propose a fast double-coupled nonnegative Canonical Polyadic Decomposition (FDC-NCPD)

Paper Details

Authors:
Tapani Ristaniemi, Fengyu Cong
Submitted On:
8 May 2019 - 6:25am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster_Xiulin Wang.pdf

(20)

Keywords

Additional Categories

Subscribe

[1] Tapani Ristaniemi, Fengyu Cong, "Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4063. Accessed: Oct. 16, 2019.
@article{4063-19,
url = {http://sigport.org/4063},
author = {Tapani Ristaniemi; Fengyu Cong },
publisher = {IEEE SigPort},
title = {Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition},
year = {2019} }
TY - EJOUR
T1 - Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition
AU - Tapani Ristaniemi; Fengyu Cong
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4063
ER -
Tapani Ristaniemi, Fengyu Cong. (2019). Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition. IEEE SigPort. http://sigport.org/4063
Tapani Ristaniemi, Fengyu Cong, 2019. Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition. Available at: http://sigport.org/4063.
Tapani Ristaniemi, Fengyu Cong. (2019). "Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition." Web.
1. Tapani Ristaniemi, Fengyu Cong. Fast Implementation of Double-coupled Nonnegative Canonical Polyadic Decomposition [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4063

Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors


This paper proposes a novel framework to regularize the highly illposed and non-linear Fourier ptychography problem using generative models. We demonstrate experimentally that our proposed algorithm, Deep Ptych, outperforms the existing Fourier ptychography techniques, in terms of quality of reconstruction and robustness against noise, using far fewer samples. We further modify the proposed approach to allow the generative model to explore solutions outside the range, leading to improved performance.

Paper Details

Authors:
Fahad Shamshad, Farwa Abbas, Ali Ahmed
Submitted On:
8 May 2019 - 3:48am
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

Deep Ptych Subsampled Fourier Ptychography Using Deep Generative Priors.pdf

(32)

Keywords

Additional Categories

Subscribe

[1] Fahad Shamshad, Farwa Abbas, Ali Ahmed, "Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4038. Accessed: Oct. 16, 2019.
@article{4038-19,
url = {http://sigport.org/4038},
author = {Fahad Shamshad; Farwa Abbas; Ali Ahmed },
publisher = {IEEE SigPort},
title = {Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors},
year = {2019} }
TY - EJOUR
T1 - Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors
AU - Fahad Shamshad; Farwa Abbas; Ali Ahmed
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4038
ER -
Fahad Shamshad, Farwa Abbas, Ali Ahmed. (2019). Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors. IEEE SigPort. http://sigport.org/4038
Fahad Shamshad, Farwa Abbas, Ali Ahmed, 2019. Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors. Available at: http://sigport.org/4038.
Fahad Shamshad, Farwa Abbas, Ali Ahmed. (2019). "Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors." Web.
1. Fahad Shamshad, Farwa Abbas, Ali Ahmed. Deep Ptych: Subsampled Fourier Ptychography Using Deep Generative Priors [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4038

Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation

Paper Details

Authors:
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li
Submitted On:
7 May 2019 - 9:56pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSPposter_常伟.pdf

(26)

Subscribe

[1] Wei Chang, Feiping Nie, Rong Wang, Xuelong Li, "Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3987. Accessed: Oct. 16, 2019.
@article{3987-19,
url = {http://sigport.org/3987},
author = {Wei Chang; Feiping Nie; Rong Wang; Xuelong Li },
publisher = {IEEE SigPort},
title = {Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation},
year = {2019} }
TY - EJOUR
T1 - Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation
AU - Wei Chang; Feiping Nie; Rong Wang; Xuelong Li
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3987
ER -
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li. (2019). Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation. IEEE SigPort. http://sigport.org/3987
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li, 2019. Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation. Available at: http://sigport.org/3987.
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li. (2019). "Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation." Web.
1. Wei Chang, Feiping Nie, Rong Wang, Xuelong Li. Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3987

Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation

Paper Details

Authors:
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li
Submitted On:
7 May 2019 - 9:56pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSPposter_常伟.pdf

(23)

Subscribe

[1] Wei Chang, Feiping Nie, Rong Wang, Xuelong Li, "Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3984. Accessed: Oct. 16, 2019.
@article{3984-19,
url = {http://sigport.org/3984},
author = {Wei Chang; Feiping Nie; Rong Wang; Xuelong Li },
publisher = {IEEE SigPort},
title = {Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation},
year = {2019} }
TY - EJOUR
T1 - Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation
AU - Wei Chang; Feiping Nie; Rong Wang; Xuelong Li
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3984
ER -
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li. (2019). Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation. IEEE SigPort. http://sigport.org/3984
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li, 2019. Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation. Available at: http://sigport.org/3984.
Wei Chang, Feiping Nie, Rong Wang, Xuelong Li. (2019). "Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation." Web.
1. Wei Chang, Feiping Nie, Rong Wang, Xuelong Li. Robust Subspace Clustering by Learning an Optimal Structured Bipartite Graph via Low-rank Representation [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3984

Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders

Paper Details

Authors:
Submitted On:
7 May 2019 - 2:36pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:

Document Files

ICASSP poster single.pdf

(22)

Subscribe

[1] , "Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3944. Accessed: Oct. 16, 2019.
@article{3944-19,
url = {http://sigport.org/3944},
author = { },
publisher = {IEEE SigPort},
title = {Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders},
year = {2019} }
TY - EJOUR
T1 - Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders
AU -
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3944
ER -
. (2019). Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders. IEEE SigPort. http://sigport.org/3944
, 2019. Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders. Available at: http://sigport.org/3944.
. (2019). "Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders." Web.
1. . Denoising Gravitational Waves with Enhanced Deep Recurrent Denoising Auto-Encoders [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3944

ACTIVE LEARNING WITH LABEL PROPORTIONS


Active Learning (AL) refers to the setting where the learner has the ability to perform queries to an oracle to acquire the true label of an instance or, sometimes, a set of instances. Even though Active Learning has been studied extensively, the setting is usually restricted to assume that the oracle is trustworthy and will provide the actual label. We argue that, while common, this approach can be made more flexible to account for different forms of supervision.

Paper Details

Authors:
Raul Santos-Rodriguez, Niall Twomey
Submitted On:
1 March 2019 - 1:22pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

empty.pdf

(96)

Subscribe

[1] Raul Santos-Rodriguez, Niall Twomey, "ACTIVE LEARNING WITH LABEL PROPORTIONS", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/3855. Accessed: Oct. 16, 2019.
@article{3855-19,
url = {http://sigport.org/3855},
author = {Raul Santos-Rodriguez; Niall Twomey },
publisher = {IEEE SigPort},
title = {ACTIVE LEARNING WITH LABEL PROPORTIONS},
year = {2019} }
TY - EJOUR
T1 - ACTIVE LEARNING WITH LABEL PROPORTIONS
AU - Raul Santos-Rodriguez; Niall Twomey
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/3855
ER -
Raul Santos-Rodriguez, Niall Twomey. (2019). ACTIVE LEARNING WITH LABEL PROPORTIONS. IEEE SigPort. http://sigport.org/3855
Raul Santos-Rodriguez, Niall Twomey, 2019. ACTIVE LEARNING WITH LABEL PROPORTIONS. Available at: http://sigport.org/3855.
Raul Santos-Rodriguez, Niall Twomey. (2019). "ACTIVE LEARNING WITH LABEL PROPORTIONS." Web.
1. Raul Santos-Rodriguez, Niall Twomey. ACTIVE LEARNING WITH LABEL PROPORTIONS [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/3855

Defending DNN Adversarial Attacks with Pruning and Logits Augmentation


Deep neural networks (DNNs) have been shown to be powerful models and perform extremely well on many complicated artificial intelligent tasks. However, recent research found that these powerful models are vulnerable to adversarial attacks, i.e., intentionally added imperceptible perturbations to DNN inputs can easily mislead the DNNs with extremely high confidence.

Paper Details

Authors:
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin
Submitted On:
28 November 2018 - 9:00pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

GlobalSip_Final.pdf

(84)

Subscribe

[1] Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin, "Defending DNN Adversarial Attacks with Pruning and Logits Augmentation", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3829. Accessed: Oct. 16, 2019.
@article{3829-18,
url = {http://sigport.org/3829},
author = {Siyue Wang; Xiao Wang; Shaokai Ye; Pu Zhao; Xue Lin },
publisher = {IEEE SigPort},
title = {Defending DNN Adversarial Attacks with Pruning and Logits Augmentation},
year = {2018} }
TY - EJOUR
T1 - Defending DNN Adversarial Attacks with Pruning and Logits Augmentation
AU - Siyue Wang; Xiao Wang; Shaokai Ye; Pu Zhao; Xue Lin
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3829
ER -
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin. (2018). Defending DNN Adversarial Attacks with Pruning and Logits Augmentation. IEEE SigPort. http://sigport.org/3829
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin, 2018. Defending DNN Adversarial Attacks with Pruning and Logits Augmentation. Available at: http://sigport.org/3829.
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin. (2018). "Defending DNN Adversarial Attacks with Pruning and Logits Augmentation." Web.
1. Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin. Defending DNN Adversarial Attacks with Pruning and Logits Augmentation [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3829

Defending DNN Adversarial Attacks with Pruning and Logits Augmentation


Deep neural networks (DNNs) have been shown to be powerful models and perform extremely well on many complicated artificial intelligent tasks. However, recent research found that these powerful models are vulnerable to adversarial attacks, i.e., intentionally added imperceptible perturbations to DNN inputs can easily mislead the DNNs with extremely high confidence.

Paper Details

Authors:
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin
Submitted On:
28 November 2018 - 8:40pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

GlobalSip_Final.pptx

(73)

Subscribe

[1] Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin, "Defending DNN Adversarial Attacks with Pruning and Logits Augmentation", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3828. Accessed: Oct. 16, 2019.
@article{3828-18,
url = {http://sigport.org/3828},
author = {Siyue Wang; Xiao Wang; Shaokai Ye; Pu Zhao; Xue Lin },
publisher = {IEEE SigPort},
title = {Defending DNN Adversarial Attacks with Pruning and Logits Augmentation},
year = {2018} }
TY - EJOUR
T1 - Defending DNN Adversarial Attacks with Pruning and Logits Augmentation
AU - Siyue Wang; Xiao Wang; Shaokai Ye; Pu Zhao; Xue Lin
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3828
ER -
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin. (2018). Defending DNN Adversarial Attacks with Pruning and Logits Augmentation. IEEE SigPort. http://sigport.org/3828
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin, 2018. Defending DNN Adversarial Attacks with Pruning and Logits Augmentation. Available at: http://sigport.org/3828.
Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin. (2018). "Defending DNN Adversarial Attacks with Pruning and Logits Augmentation." Web.
1. Siyue Wang, Xiao Wang, Shaokai Ye, Pu Zhao, Xue Lin. Defending DNN Adversarial Attacks with Pruning and Logits Augmentation [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3828

Pages