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

Other applications of machine learning (MLR-APPL)

Nonnegative Matrix Factorization with Transform Learning


Traditional NMF-based signal decomposition relies on the factorization of spectral data, which is typically computed by means of short-time frequency transform. In this paper we propose to relax the choice of a pre-fixed transform and learn a short-time orthogonal transform together with the factorization. To this end, we formulate a regularized optimization problem reminiscent of conventional NMF, yet with the transform as additional unknown parameters, and design a novel block-descent algorithm enabling to find stationary points of this objective function.

Paper Details

Authors:
Dylan Fagot, Herwig Wendt, Cédric Févotte
Submitted On:
12 April 2018 - 4:53pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster_landscape.pdf

(119)

Keywords

Additional Categories

Subscribe

[1] Dylan Fagot, Herwig Wendt, Cédric Févotte, "Nonnegative Matrix Factorization with Transform Learning", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2501. Accessed: Jul. 23, 2019.
@article{2501-18,
url = {http://sigport.org/2501},
author = {Dylan Fagot; Herwig Wendt; Cédric Févotte },
publisher = {IEEE SigPort},
title = {Nonnegative Matrix Factorization with Transform Learning},
year = {2018} }
TY - EJOUR
T1 - Nonnegative Matrix Factorization with Transform Learning
AU - Dylan Fagot; Herwig Wendt; Cédric Févotte
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2501
ER -
Dylan Fagot, Herwig Wendt, Cédric Févotte. (2018). Nonnegative Matrix Factorization with Transform Learning. IEEE SigPort. http://sigport.org/2501
Dylan Fagot, Herwig Wendt, Cédric Févotte, 2018. Nonnegative Matrix Factorization with Transform Learning. Available at: http://sigport.org/2501.
Dylan Fagot, Herwig Wendt, Cédric Févotte. (2018). "Nonnegative Matrix Factorization with Transform Learning." Web.
1. Dylan Fagot, Herwig Wendt, Cédric Févotte. Nonnegative Matrix Factorization with Transform Learning [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2501

OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS


We present a new method to generate fake data in unknown classes in generative adversarial networks (GANs) framework. The generator in GANs is trained to generate somewhat similar to data in known classes but the different one by modelling noisy distribution on feature space of a classifier using proposed marginal denoising autoencoder. The generated data are treated as fake instances in unknown classes and given to the classifier to make it be robust to the real unknown classes.

Paper Details

Authors:
Inhyuk Jo, Jungtaek Kim, Hyohyeong Kang, Yong-Deok Kim, Seungjin Choi
Submitted On:
12 April 2018 - 4:21pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

poster_icassp_2018_b.pdf

(205)

Subscribe

[1] Inhyuk Jo, Jungtaek Kim, Hyohyeong Kang, Yong-Deok Kim, Seungjin Choi, "OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2494. Accessed: Jul. 23, 2019.
@article{2494-18,
url = {http://sigport.org/2494},
author = {Inhyuk Jo; Jungtaek Kim; Hyohyeong Kang; Yong-Deok Kim; Seungjin Choi },
publisher = {IEEE SigPort},
title = {OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS},
year = {2018} }
TY - EJOUR
T1 - OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS
AU - Inhyuk Jo; Jungtaek Kim; Hyohyeong Kang; Yong-Deok Kim; Seungjin Choi
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2494
ER -
Inhyuk Jo, Jungtaek Kim, Hyohyeong Kang, Yong-Deok Kim, Seungjin Choi. (2018). OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS. IEEE SigPort. http://sigport.org/2494
Inhyuk Jo, Jungtaek Kim, Hyohyeong Kang, Yong-Deok Kim, Seungjin Choi, 2018. OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS. Available at: http://sigport.org/2494.
Inhyuk Jo, Jungtaek Kim, Hyohyeong Kang, Yong-Deok Kim, Seungjin Choi. (2018). "OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS." Web.
1. Inhyuk Jo, Jungtaek Kim, Hyohyeong Kang, Yong-Deok Kim, Seungjin Choi. OPEN SET RECOGNITION BY REGULARISING CLASSIFIER WITH FAKE DATA GENERATED BY GENERATIVE ADVERSARIAL NETWORKS [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2494

A feature fusion method based on extreme learning machine for speech emotion recognition


Speech emotion recognition is important to understand users' intention in human-computer interaction. However, it is a challenging task partly because we cannot clearly know which feature and model are effective to distinguish emotions. Previous studies utilize convolutional neural network (CNN) directly on spectrograms to extract features, and bidirectional long short term memory (BLSTM) is the state-of-the-art model. However, there are two problems of CNN-BLSTM. Firstly, it doesn't utilize heuristic features based on priori knowledge.

Paper Details

Authors:
Longbiao Wang, Jianwu Dang, Linjuan Zhang, Haotian Guan
Submitted On:
12 April 2018 - 12:07pm
Short Link:
Type:
Event:
Paper Code:
Document Year:
Cite

Document Files

poster-guolili.pdf

(249)

Subscribe

[1] Longbiao Wang, Jianwu Dang, Linjuan Zhang, Haotian Guan, "A feature fusion method based on extreme learning machine for speech emotion recognition", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2426. Accessed: Jul. 23, 2019.
@article{2426-18,
url = {http://sigport.org/2426},
author = {Longbiao Wang; Jianwu Dang; Linjuan Zhang; Haotian Guan },
publisher = {IEEE SigPort},
title = {A feature fusion method based on extreme learning machine for speech emotion recognition},
year = {2018} }
TY - EJOUR
T1 - A feature fusion method based on extreme learning machine for speech emotion recognition
AU - Longbiao Wang; Jianwu Dang; Linjuan Zhang; Haotian Guan
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2426
ER -
Longbiao Wang, Jianwu Dang, Linjuan Zhang, Haotian Guan. (2018). A feature fusion method based on extreme learning machine for speech emotion recognition. IEEE SigPort. http://sigport.org/2426
Longbiao Wang, Jianwu Dang, Linjuan Zhang, Haotian Guan, 2018. A feature fusion method based on extreme learning machine for speech emotion recognition. Available at: http://sigport.org/2426.
Longbiao Wang, Jianwu Dang, Linjuan Zhang, Haotian Guan. (2018). "A feature fusion method based on extreme learning machine for speech emotion recognition." Web.
1. Longbiao Wang, Jianwu Dang, Linjuan Zhang, Haotian Guan. A feature fusion method based on extreme learning machine for speech emotion recognition [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2426

Performance Benchmarks for Detection Problems


We propose a benchmark curve that measures the inherent complexity of a detection problem. The benchmark curve is built using a sequence of simple detection methods based upon random projection. It is parameterized by the area above the receiver-operating characteristic curve of the detection method and its computational cost. It divides the plane into regions that can be used to characterize the computational and structural advantages of a given detection method. Numerical illustrations are provided.

Paper Details

Authors:
Kelsie Larson, Mireille Boutin
Submitted On:
13 November 2017 - 12:58am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Performance Benchmarks for Detection Problems

(162)

Subscribe

[1] Kelsie Larson, Mireille Boutin, "Performance Benchmarks for Detection Problems", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2332. Accessed: Jul. 23, 2019.
@article{2332-17,
url = {http://sigport.org/2332},
author = {Kelsie Larson; Mireille Boutin },
publisher = {IEEE SigPort},
title = {Performance Benchmarks for Detection Problems},
year = {2017} }
TY - EJOUR
T1 - Performance Benchmarks for Detection Problems
AU - Kelsie Larson; Mireille Boutin
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2332
ER -
Kelsie Larson, Mireille Boutin. (2017). Performance Benchmarks for Detection Problems. IEEE SigPort. http://sigport.org/2332
Kelsie Larson, Mireille Boutin, 2017. Performance Benchmarks for Detection Problems. Available at: http://sigport.org/2332.
Kelsie Larson, Mireille Boutin. (2017). "Performance Benchmarks for Detection Problems." Web.
1. Kelsie Larson, Mireille Boutin. Performance Benchmarks for Detection Problems [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2332

A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T

Paper Details

Authors:
Submitted On:
10 November 2017 - 9:03am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

globalSIP.pdf

(2119)

Subscribe

[1] , "A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2291. Accessed: Jul. 23, 2019.
@article{2291-17,
url = {http://sigport.org/2291},
author = { },
publisher = {IEEE SigPort},
title = {A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T},
year = {2017} }
TY - EJOUR
T1 - A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T
AU -
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2291
ER -
. (2017). A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T. IEEE SigPort. http://sigport.org/2291
, 2017. A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T. Available at: http://sigport.org/2291.
. (2017). "A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T." Web.
1. . A 200MHZ 202.4GFLOPS@10.8W VGG16 ACCELERATOR IN XILINX VX690T [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2291

Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing

Paper Details

Authors:
Danfeng Hong, Naoto Yokoya, Jocelyn Chanussot, Xiao Xiang Zhu
Submitted On:
16 September 2017 - 4:51am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

hyperspectral unmixing

(239)

Keywords

Additional Categories

Subscribe

[1] Danfeng Hong, Naoto Yokoya, Jocelyn Chanussot, Xiao Xiang Zhu, "Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2184. Accessed: Jul. 23, 2019.
@article{2184-17,
url = {http://sigport.org/2184},
author = {Danfeng Hong; Naoto Yokoya; Jocelyn Chanussot; Xiao Xiang Zhu },
publisher = {IEEE SigPort},
title = {Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing},
year = {2017} }
TY - EJOUR
T1 - Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing
AU - Danfeng Hong; Naoto Yokoya; Jocelyn Chanussot; Xiao Xiang Zhu
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2184
ER -
Danfeng Hong, Naoto Yokoya, Jocelyn Chanussot, Xiao Xiang Zhu. (2017). Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing. IEEE SigPort. http://sigport.org/2184
Danfeng Hong, Naoto Yokoya, Jocelyn Chanussot, Xiao Xiang Zhu, 2017. Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing. Available at: http://sigport.org/2184.
Danfeng Hong, Naoto Yokoya, Jocelyn Chanussot, Xiao Xiang Zhu. (2017). "Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing." Web.
1. Danfeng Hong, Naoto Yokoya, Jocelyn Chanussot, Xiao Xiang Zhu. Learning a Low-coherence Dictionary to Address Spectral Variability for Hyperspectral Unmixing [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2184

Greedy Deep Transform Learning


We introduce deep transform learning – a new
tool for deep learning. Deeper representation is learnt by
stacking one transform after another. The learning proceeds in
a greedy way. The first layer learns the transform and features
from the input training samples. Subsequent layers use the
features (after activation) from the previous layers as training
input. Experiments have been carried out with other deep
representation learning tools – deep dictionary learning,
stacked denoising autoencoder, deep belief network and PCANet

Paper Details

Authors:
Jyoti Maggu, Angshul Majumdar
Submitted On:
18 September 2017 - 1:57pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICIP_greedyDTL.pdf

(294)

Subscribe

[1] Jyoti Maggu, Angshul Majumdar, "Greedy Deep Transform Learning", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/2180. Accessed: Jul. 23, 2019.
@article{2180-17,
url = {http://sigport.org/2180},
author = {Jyoti Maggu; Angshul Majumdar },
publisher = {IEEE SigPort},
title = {Greedy Deep Transform Learning},
year = {2017} }
TY - EJOUR
T1 - Greedy Deep Transform Learning
AU - Jyoti Maggu; Angshul Majumdar
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/2180
ER -
Jyoti Maggu, Angshul Majumdar. (2017). Greedy Deep Transform Learning. IEEE SigPort. http://sigport.org/2180
Jyoti Maggu, Angshul Majumdar, 2017. Greedy Deep Transform Learning. Available at: http://sigport.org/2180.
Jyoti Maggu, Angshul Majumdar. (2017). "Greedy Deep Transform Learning." Web.
1. Jyoti Maggu, Angshul Majumdar. Greedy Deep Transform Learning [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/2180

Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation


This paper presents preliminary results for motion behavior analysis of Madagascar hissing cockroach biobots subject to stochastic and periodic neurostimulation pulses corresponding to randomly applied right and left turn, and move forward commands. We present our experimental setup and propose an unguided search strategy based stimulation profile designed for exploration of unknown environments. We study a probabilistic motion model fitted to the trajectories of biobots, perturbed from their natural motion by the stimulation pulses.

Paper Details

Authors:
Alireza Dirafzoon, Tahmid Latif, Fengyuan Gong, Mihail Sichitiu, Alper Bozkurt, Edgar Lobaton
Submitted On:
6 March 2017 - 7:40am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Poster.pdf

(1120)

Subscribe

[1] Alireza Dirafzoon, Tahmid Latif, Fengyuan Gong, Mihail Sichitiu, Alper Bozkurt, Edgar Lobaton, "Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1646. Accessed: Jul. 23, 2019.
@article{1646-17,
url = {http://sigport.org/1646},
author = {Alireza Dirafzoon; Tahmid Latif; Fengyuan Gong; Mihail Sichitiu; Alper Bozkurt; Edgar Lobaton },
publisher = {IEEE SigPort},
title = {Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation},
year = {2017} }
TY - EJOUR
T1 - Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation
AU - Alireza Dirafzoon; Tahmid Latif; Fengyuan Gong; Mihail Sichitiu; Alper Bozkurt; Edgar Lobaton
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1646
ER -
Alireza Dirafzoon, Tahmid Latif, Fengyuan Gong, Mihail Sichitiu, Alper Bozkurt, Edgar Lobaton. (2017). Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation. IEEE SigPort. http://sigport.org/1646
Alireza Dirafzoon, Tahmid Latif, Fengyuan Gong, Mihail Sichitiu, Alper Bozkurt, Edgar Lobaton, 2017. Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation. Available at: http://sigport.org/1646.
Alireza Dirafzoon, Tahmid Latif, Fengyuan Gong, Mihail Sichitiu, Alper Bozkurt, Edgar Lobaton. (2017). "Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation." Web.
1. Alireza Dirafzoon, Tahmid Latif, Fengyuan Gong, Mihail Sichitiu, Alper Bozkurt, Edgar Lobaton. Biobotic Motion and Behavior Analysis in Response to Directional Neurostimulation [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1646

mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis

Paper Details

Authors:
Rui Zhang, Feiping Nie, Xuelong Li
Submitted On:
2 March 2017 - 4:13pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

beamer of EC

(74)

Keywords

Additional Categories

Subscribe

[1] Rui Zhang, Feiping Nie, Xuelong Li, "mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1596. Accessed: Jul. 23, 2019.
@article{1596-17,
url = {http://sigport.org/1596},
author = {Rui Zhang; Feiping Nie; Xuelong Li },
publisher = {IEEE SigPort},
title = {mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis},
year = {2017} }
TY - EJOUR
T1 - mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis
AU - Rui Zhang; Feiping Nie; Xuelong Li
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1596
ER -
Rui Zhang, Feiping Nie, Xuelong Li. (2017). mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis. IEEE SigPort. http://sigport.org/1596
Rui Zhang, Feiping Nie, Xuelong Li, 2017. mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis. Available at: http://sigport.org/1596.
Rui Zhang, Feiping Nie, Xuelong Li. (2017). "mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis." Web.
1. Rui Zhang, Feiping Nie, Xuelong Li. mbedded Clustering via Robust Orthogonal Least Square Discriminant Analysis [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1596

A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT


Practical limitations on the duration of individual fMRI scans have led neuroscientist to consider the aggregation of data from multiple subjects. Differences in anatomical structures and functional topographies of brains require aligning data across subjects. Existing functional alignment methods serve as a preprocessing step that allows subsequent statistical methods to learn from the aggregated multi-subject data. Despite their success, current alignment methods do not leverage the labeled data used in the subsequent methods.

Paper Details

Authors:
Javier S. Turek, Theodore L. Willke, Po-Hsuan Chen, Peter J. Ramadge
Submitted On:
2 March 2017 - 12:56pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Semi-Supervised fMRI Functional Alignment

(258)

Subscribe

[1] Javier S. Turek, Theodore L. Willke, Po-Hsuan Chen, Peter J. Ramadge, "A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1587. Accessed: Jul. 23, 2019.
@article{1587-17,
url = {http://sigport.org/1587},
author = {Javier S. Turek; Theodore L. Willke; Po-Hsuan Chen; Peter J. Ramadge },
publisher = {IEEE SigPort},
title = {A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT},
year = {2017} }
TY - EJOUR
T1 - A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT
AU - Javier S. Turek; Theodore L. Willke; Po-Hsuan Chen; Peter J. Ramadge
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1587
ER -
Javier S. Turek, Theodore L. Willke, Po-Hsuan Chen, Peter J. Ramadge. (2017). A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT. IEEE SigPort. http://sigport.org/1587
Javier S. Turek, Theodore L. Willke, Po-Hsuan Chen, Peter J. Ramadge, 2017. A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT. Available at: http://sigport.org/1587.
Javier S. Turek, Theodore L. Willke, Po-Hsuan Chen, Peter J. Ramadge. (2017). "A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT." Web.
1. Javier S. Turek, Theodore L. Willke, Po-Hsuan Chen, Peter J. Ramadge. A SEMI-SUPERVISED METHOD FOR MULTI-SUBJECT FMRI FUNCTIONAL ALIGNMENT [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1587

Pages