Transfer Learning-Based Independent Component Analysis

Zheng, Ziqian; Zhao, Wei; Hable, Brock; Gong, Yutao; Wang, Xuan; Shannon, Robert W.; Liu, Kaibo

doi:10.1109/tase.2022.3229294

Title: Transfer Learning-Based Independent Component Analysis

Journal Article · Tue Dec 20 00:00:00 EST 2022 · IEEE Transactions on Automation Science and Engineering

DOI:https://doi.org/10.1109/tase.2022.3229294· OSTI ID:2007823

^[1]; Zhao, Wei ^[2]; Hable, Brock ^[2];

^[2]; Wang, Xuan ^[2]; Shannon, Robert W. ^[2];

^[1]

Univ. of Wisconsin, Madison, WI (United States)
3M Company, Maplewood, MN (United States)

Understanding the underlying component structure is crucial for multivariate signal analysis. Among all the techniques that try to learn the latent structure, independent component analysis (ICA) is one of the most important and popular methods, which aims to extract independent components from multivariate signals and enables further analysis. For example, in electroencephalogram (EEG) analysis, artifacts filtering and disease detection are conducted based on the independent components of the signals. One critical challenge in existing ICA approaches is that the component extraction accuracy may degrade when the available data of a unit are limited. To address this issue, this paper proposes a transfer learning-based ICA method by innovatively transferring component distribution from a source domain, so that accurate component extraction results can be achieved even when only limited data are available in the target domain. To the best of our knowledge, this is the first work that leverages transfer learning to improve ICA accuracy with limited available data. In particular, we first extract all the independent components from the source domain by maximizing the log-likelihood function with a Newton-like method on a smooth manifold. Then for the target domain, the component with the largest negentropy is extracted in each round. To effectively leverage the knowledge from the source domain and to prevent the negative transfer, we try to find a component in the source domain that matches the component we are extracting. The probability density function of the matched component will then be used to improve the component extraction accuracy if such matched component can be found; otherwise, no knowledge will be transferred. Finally, numerical simulations and a case study with electrocardiogram (ECG) data are conducted, showing the effectiveness of the proposed method in transferring knowledge and reducing negative transfer.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: Univ. of Wisconsin, Madison, WI (United States)

Sponsoring Organization:: USDOE

Grant/Contract Number:: NE0008993

OSTI ID:: 2007823

Journal Information:: IEEE Transactions on Automation Science and Engineering, Vol. 21, Issue 1; ISSN 1545-5955

Publisher:: IEEECopyright Statement

Country of Publication:: United States

Language:: English

References (31)

Statistical degradation modeling and prognostics of multiple sensor signals via data fusion: A composite health index approach Song, Changyue; Liu, Kaibo IISE Transactions, Vol. 50, Issue 10 https://doi.org/10.1080/24725854.2018.1440673	journal	May 2018
A Generic Health Index Approach for Multisensor Degradation Modeling and Sensor Selection Kim, Minhee; Song, Changyue; Liu, Kaibo IEEE Transactions on Automation Science and Engineering, Vol. 16, Issue 3 https://doi.org/10.1109/TASE.2018.2890608	journal	July 2019
An Improved FastICA Method for Fetal ECG Extraction Yuan, Li; Zhou, Zhuhuang; Yuan, Yanchao Computational and Mathematical Methods in Medicine, Vol. 2018 https://doi.org/10.1155/2018/7061456	journal	January 2018
Independent component analysis of functional MRI: what is signal and what is noise? McKeown, M. Current Opinion in Neurobiology, Vol. 13, Issue 5 https://doi.org/10.1016/j.conb.2003.09.012	journal	October 2003
Process Monitoring Based on Independent Component Analysis−Principal Component Analysis (ICA−PCA) and Similarity Factors Ge, Zhiqiang; Song, Zhihuan Industrial & Engineering Chemistry Research, Vol. 46, Issue 7 https://doi.org/10.1021/ie061083g	journal	March 2007
Data-driven human transcriptomic modules determined by independent component analysis Zhou, Weizhuang; Altman, Russ B. BMC Bioinformatics, Vol. 19, Issue 1 https://doi.org/10.1186/s12859-018-2338-4	journal	September 2018
A denoising carbon price forecasting method based on the integration of kernel independent component analysis and least squares support vector regression E., Jianwei; Ye, Jimin; He, Lulu Neurocomputing, Vol. 434 https://doi.org/10.1016/j.neucom.2020.12.086	journal	April 2021
Infomax and maximum likelihood for blind source separation Cardoso, J. -F. IEEE Signal Processing Letters, Vol. 4, Issue 4 https://doi.org/10.1109/97.566704	journal	April 1997
High-Order Contrasts for Independent Component Analysis Cardoso, Jean-François Neural Computation, Vol. 11, Issue 1 https://doi.org/10.1162/089976699300016863	journal	January 1999
Blind signal separation: statistical principles Cardoso, J. -F. Proceedings of the IEEE, Vol. 86, Issue 10 https://doi.org/10.1109/5.720250	journal	January 1998
Independent Component Analysis Based on Nonparametric Density Estimation Boscolo, R.; Pan, H.; Roychowdhury, V. P. IEEE Transactions on Neural Networks, Vol. 15, Issue 1 https://doi.org/10.1109/TNN.2003.820667	journal	January 2004
Blind separation of mixture of independent sources through a quasi-maximum likelihood approach IEEE Transactions on Signal Processing, Vol. 45, Issue 7 https://doi.org/10.1109/78.599941	journal	July 1997
Independent component analysis via nonparametric maximum likelihood estimation Samworth, Richard J.; Yuan, Ming The Annals of Statistics, Vol. 40, Issue 6 https://doi.org/10.1214/12-AOS1060	journal	December 2012
Fast Kernel-Based Independent Component Analysis IEEE Transactions on Signal Processing, Vol. 57, Issue 9 https://doi.org/10.1109/TSP.2009.2022857	journal	September 2009
A Survey on Transfer Learning Pan, Sinno Jialin; Yang, Qiang IEEE Transactions on Knowledge and Data Engineering, Vol. 22, Issue 10 https://doi.org/10.1109/TKDE.2009.191	journal	October 2010
Boosting for transfer learning Dai, Wenyuan; Yang, Qiang; Xue, Gui-Rong Proceedings of the 24th international conference on Machine learning - ICML '07 https://doi.org/10.1145/1273496.1273521	conference	January 2007
Boosting for transfer learning with multiple sources Yao, Yi; Doretto, Gianfranco 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.2010.5539857	conference	June 2010
Domain Adaptation via Transfer Component Analysis Pan, Sinno Jialin; Tsang, Ivor W.; Kwok, James T. IEEE Transactions on Neural Networks, Vol. 22, Issue 2 https://doi.org/10.1109/TNN.2010.2091281	journal	February 2011
Simultaneous Deep Transfer Across Domains and Tasks Tzeng, Eric; Hoffman, Judy; Darrell, Trevor 2015 IEEE International Conference on Computer Vision (ICCV) https://doi.org/10.1109/ICCV.2015.463	conference	December 2015
Deep visual domain adaptation: A survey Wang, Mei; Deng, Weihong Neurocomputing, Vol. 312 https://doi.org/10.1016/j.neucom.2018.05.083	journal	October 2018
Deep Learning for Emotion Recognition on Small Datasets using Transfer Learning Ng, Hong-Wei; Nguyen, Viet Dung; Vonikakis, Vassilios Proceedings of the 2015 ACM on International Conference on Multimodal Interaction https://doi.org/10.1145/2818346.2830593	conference	November 2015
Convolutional Neural Networks for Medical Image Analysis: Full Training or Fine Tuning? Tajbakhsh, Nima; Shin, Jae Y.; Gurudu, Suryakanth R. IEEE Transactions on Medical Imaging, Vol. 35, Issue 5 https://doi.org/10.1109/TMI.2016.2535302	journal	May 2016
Joint Fine-Tuning in Deep Neural Networks for Facial Expression Recognition Jung, Heechul; Lee, Sihaeng; Yim, Junho 2015 IEEE International Conference on Computer Vision (ICCV) https://doi.org/10.1109/ICCV.2015.341	conference	December 2015
Independent component analysis: An introduction Tharwat, Alaa Applied Computing and Informatics, Vol. 17, Issue 2 https://doi.org/10.1016/j.aci.2018.08.006	journal	August 2020
A Review of Independent Component Analysis Techniques and their Applications Acharya, Dp; Panda, G. IETE Technical Review, Vol. 25, Issue 6 https://doi.org/10.4103/0256-4602.45424	journal	January 2008
Density Estimation for Statistics and Data Analysis Silverman, Bernard W. https://doi.org/10.1201/9781315140919	book	January 1998
On the entropy minimization of a linear mixture of variables for source separation Vrins, Frédéric; Verleysen, Michel Signal Processing, Vol. 85, Issue 5 https://doi.org/10.1016/j.sigpro.2004.11.026	journal	May 2005
Fast ICA based technique for non-invasive fetal ECG extraction Nikam, Sonal; Deosarkar, Shankar 2016 Conference on Advances in Signal Processing (CASP) https://doi.org/10.1109/CASP.2016.7746138	conference	June 2016
Non-invasive Fetal Electrocardiography for Intrapartum Cardiotocography Vullings, Rik; van Laar, Judith O. E. H. Frontiers in Pediatrics, Vol. 8 https://doi.org/10.3389/fped.2020.599049	journal	December 2020
Determination of fetal heart rate from abdominal signals: evaluation of beat-to-beat accuracy in relation to the direct fetal electrocardiogram Jezewski, Janusz; Matonia, Adam; Kupka, Tomasz Biomedizinische Technik/Biomedical Engineering, Vol. 57, Issue 5 https://doi.org/10.1515/bmt-2011-0130	journal	January 2012
Misidentification of Maternal Heart Rate as Fetal on Cardiotocography During the Second Stage of Labor Nurani, Raisha; Chandraharan, Edwin; Lowe, Virginia Obstetrical & Gynecological Survey, Vol. 68, Issue 4 https://doi.org/10.1097/01.ogx.0000429288.27395.51	journal	April 2013

Similar Records

Towards intelligent diagnostic system employing integration of mathematical and engineering model

Journal Article · Fri May 15 00:00:00 EDT 2015 · AIP Conference Proceedings · OSTI ID:2007823

Isa, Nor Ashidi Mat

Structured Prompt Interrogation and Recursive Extraction of Semantics (SPIRES): a method for populating knowledge bases using zero-shot learning

Journal Article · Wed Feb 21 00:00:00 EST 2024 · Bioinformatics · OSTI ID:2007823

Caufield, J. Harry; Hegde, Harshad; Emonet, Vincent; +10 more

Imaging of Calibrated Defects in Additively Manufactured Materials: Pulsed Thermal Tomography Nondestructive Examination of Additively Manufactured Reactor Materials and Components

Technical Report · Tue Mar 31 00:00:00 EDT 2020 · OSTI ID:2007823

Heifetz, Alexander; Zhang, Xin; Saniie, Jafar; +5 more

Related Subjects

ICA
independent component matching
limited data
negative transfer
transfer learning
task analysis
Kernel
electrocardiography
electroencephalography
data models

Title: Transfer Learning-Based Independent Component Analysis

Citation Formats

References (31)

Similar Records

Related Subjects