research-article

Open access

Multi-Source Deep Domain Adaptation with Weak Supervision for Time-Series Sensor Data

Authors:

Garrett Wilson,

Janardhan Rao Doppa,

Diane J. CookAuthors Info & Claims

KDD '20: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

Pages 1768 - 1778

https://doi.org/10.1145/3394486.3403228

Published: 20 August 2020 Publication History

Abstract

Domain adaptation (DA) offers a valuable means to reuse data and models for new problem domains. However, robust techniques have not yet been considered for time series data with varying amounts of data availability. In this paper, we make three main contributions to fill this gap. First, we propose a novel Convolutional deep Domain Adaptation model for Time Series data (CoDATS) that significantly improves accuracy and training time over state-of-the-art DA strategies on real-world sensor data benchmarks. By utilizing data from multiple source domains, we increase the usefulness of CoDATS to further improve accuracy over prior single-source methods, particularly on complex time series datasets that have high variability between domains. Second, we propose a novel Domain Adaptation with Weak Supervision (DA-WS) method by utilizing weak supervision in the form of target-domain label distributions, which may be easier to collect than additional data labels. Third, we perform comprehensive experiments on diverse real-world datasets to evaluate the effectiveness of our domain adaptation and weak supervision methods. Results show that CoDATS for single-source DA significantly improves over the state-of-the-art methods, and we achieve additional improvements in accuracy using data from multiple source domains and weakly supervised signals.

Supplementary Material

MP4 File (3394486.3403228.mp4)

Video presentation of our paper "Multi-Source Deep Domain Adaptation with Weak Supervision for Time-Series Sensor Data"

Download
587.61 MB

References

[1]

D. Anguita, A. Ghio, L. Oneto, X. Parra, and J. L. Reyes-Ortiz. 2013. A public domain dataset for human activity recognition using smartphones. In ESANN.

[2]

S. Bai, J. Z. Kolter, et al. 2018. An empirical evaluation of generic convolutional and recurrent networks for sequence modeling. arXiv:1803.01271(2018).

[3]

S. Ben-David et al.2010. A theory of learning from different domains. Machine Learning 79, 1 (2010), 151--175.

Digital Library

[4]

K. Bousmalis et al. 2017. Unsupervised Pixel-Level Domain Adaptation With Generative Adversarial Networks. In CVPR.

[5]

C. Chen, Y. Miao, C. X. Lu, and Xie. 2019. Motion Transformer: Transferring Neural Inertial Tracking between Domains. In AAAI. 8009--8016.

[6]

H. I. Fawaz, G. Forestier, J. Weber, L. Idoumghar, and P. Muller. 2018. Transfer learning for time series classification. In IEEE Big Data. 1367--1376.

[7]

H. I. Fawaz, G. Forestier, J. Weber, L. Idoumghar, and P. Muller. 2019. Deep learning for time series classification: a review. DMKD 33, 4 (2019), 917--963.

Digital Library

[8]

H. I. Fawaz, B. Lucas, G. Forestier, C. Pelletier, et al.2019. Inception Time: Finding AlexNet for Time Series Classification. arXiv:1909.04939(2019).

[9]

G. French et al. 2018. Self-ensembling for visual domain adaptation. In ICLR.

[10]

J. Froehlich et al.2007. MyExperience: A System for in Situ Tracing and Capturing of User Feedback on Mobile Phones. In MobiSys. 57--70.

[11]

K. Ganchev, J. Gillenwater, B. Taskar, et al.2010. Posterior regularization for structured latent variable models. JMLR 11, Jul (2010), 2001--2049.

Digital Library

[12]

Y. Ganin et al.2016. Domain-Adversarial Training of Neural Networks. JMLR 17, 59 (2016), 1--35.

Digital Library

[13]

K. Greff et al. 2017. LSTM: A Search Space Odyssey.IEEE Tr. Neur. Net. Learn. Sys. 28, 10 (2017), 2222--2232.

[14]

J. Hoffman et al.2018. CyCADA: Cycle-Consistent Adversarial Domain Adaptation. In ICML, Vol. 80. 1994--2003.

[15]

S. Hong, J. Oh, H. Lee, and B. Han. 2016. Learning Transferrable Knowledge for Semantic Segmentation With Deep Convolutional Neural Network. In CVPR.

[16]

W. Hong, Z. Wang, M. Yang, and J. Yuan. 2018. Conditional Generative Adversarial Network for Structured Domain Adaptation. In CVPR.

[17]

E. Hosseini-Asl, Y. Zhou, C. Xiong, and R. Socher. 2019. Augmented Cyclic Adversarial Learning for Low Resource Domain Adaptation. In ICLR.

[18]

N. Hu, G. Englebienne, Z. Lou, et al.2017. Learning to Recognize Human Activities Using Soft Labels.IEEE Trans. Pat. Anal. Mach. Intell. 39, 10 (2017), 1973--1984.

Digital Library

[19]

F. Huang and A. Yates. 2012. Biased Representation Learning for Domain Adaptation. In EMNLP-CoNLL. 1313--1323.

[20]

W. Jiang, C. Miao, F. Ma, S. Yao, Y. Wang, Y. Yuan, et al.2018. Towards environment independent device free human activity recognition. In MobiCom. 289--304.

[21]

R. Jozefowicz, W. Zaremba, and I. Sutskever. 2015. An empirical exploration of recurrent network architectures. In ICML. 2342--2350.

[22]

G. Kang, L. Jiang, Y. Yang, and A. G. Hauptmann. 2019. Contrastive Adaptation Network for Unsupervised Domain Adaptation. arXiv:1901.00976(2019).

[23]

A. Kumar et al.2018. Co-regularized Alignment for Unsupervised Domain Adaptation. In NeurIPS 31. 9345--9356.

[24]

J. R. Kwapisz, G. M. Weiss, and S. A. Moore. 2011. Activity Recognition Using Cell Phone Accelerometers. SIGKDD Explor. Newsl.12, 2 (2011), 74--82.

Digital Library

[25]

Y. Li, N. Wang, J. Shi, X. Hou, and J. Liu. 2018. Adaptive Batch Normalization for practical domain adaptation.Pattern Recognition 80 (2018), 109 -- 117.

[26]

J. Liu, L. Zhong, and J. Wickramasuriya. 2009. uWave: Accelerometer-based personalized gesture recognition and its applications. In PerCom. 657 -- 675.

[27]

G. Melis, C. Dyer, and P. Blunsom. 2017. On the state of the art of evaluation in neural language models.arXiv:1707.05589(2017).

[28]

J. Miller and M. Hardt. 2018. Stable recurrent models. arXiv:1805.10369(2018).

[29]

B. D. Minor, J. R. Doppa, et al.2017. Learning Activity Predictors from Sensor Data:Algorithms, Evaluation, and Applications. IEEE TKDE 29, 12 (2017), 2744--2757.

[30]

R. Pascanu, T. Mikolov, and Y. Bengio. 2013. On the difficulty of training recurrent neural networks. In ICML. 1310--1318.

[31]

D. Pathak, P. Krahenbuhl, and T. Darrell. 2015. Constrained Convolutional Neural Networks for Weakly Supervised Segmentation. In ICCV.

[32]

S. Purushotham, W. Carvalho, T. Nilanon, and Y. Liu. 2017. Variational adversarial deep domain adaptation for health care time series analysis. In ICLR.

[33]

S. Sankaranarayanan, Y. Balaji, C. D. Castillo, and R. Chellappa. 2018. Generate to Adapt: Aligning Domains Using Generative Adversarial Networks. In CVPR.

[34]

R. Shu, H. Bui, H. Narui, and S. Ermon. 2018. A DIRT-T Approach to Unsupervised Domain Adaptation. In ICLR.

[35]

A. Stisen et al.2015. Smart Devices Are Different: Assessing and Mitigating Mobile Sensing Heterogeneities for Activity Recognition. In SenSys. 127--140.

[36]

S. Sun, H. Shi, and Y. Wu. 2015. A survey of multi-source domain adaptation. Information Fusion24 (2015), 84 -- 92.

[37]

M. Tonutti, E. Ruffaldi, A. Cattaneo, et al.2019. Robust and subject-independent driving manoeuvre anticipation through Domain-Adversarial Recurrent Neural Networks. Robotics and Autonomous Systems115 (2019), 162 -- 173.

[38]

V. Vercruyssen, W. Meert, and J. Davis. 2017. Transfer learning for time series anomaly detection. In CEUR Workshop Proceedings, Vol. 1924. 27--37.

[39]

Z. Wang, W. Yan, and T. Oates. 2017. Time series classification from scratch with deep neural networks: A strong baseline. In IJCNN. 1578--1585.

[40]

G. Wilson and D. J. Cook. 2019. A Survey of Unsupervised Deep Domain Adaptation. arXiv: 1812.02849(2019).

[41]

Q. Xie, Z. Dai, Y. Du, E. Hovy, and G. Neubig. 2017. Controllable In variance through Adversarial Feature Learning. In NeurIPS 30. 585--596.

[42]

H. Zhao, S. Zhang, G. Wu, J. M. F. Moura, J. P. Costeira, and G. J. Gordon. 2018. Adversarial Multiple Source Domain Adaptation. In NeurIPS 31. 8559--8570.

[43]

M. Zhao et al.2017. Learning Sleep Stages from Radio Signals: A Conditional Adversarial Architecture. In ICML, Vol. 70. 4100--4109.

Cited By

Singh AChakraborty SHe ZPang YZhang SSubedi RLustria MCharness NBoot W(2024)Predicting Adherence to Computer-Based Cognitive Training Programs Among Older Adults: Study of Domain Adaptation and Deep LearningJMIR Aging10.2196/537937(e53793-e53793)Online publication date: 16-Sep-2024
https://doi.org/10.2196/53793
Oshima HIshizone THiguchi T(2024)Inter-seasons and Inter-households Domain Adaptation Based on DANNs and Pseudo Labeling for Non-Intrusive Occupancy Detection非侵入型在宅推定に対するDANNsと疑似ラベリングをもとにした季節間および世帯間の教師なしドメイン適応手法Transactions of the Japanese Society for Artificial Intelligence10.1527/tjsai.39-5_E-O4139:5(E-O41_1-13)Online publication date: 1-Sep-2024
https://doi.org/10.1527/tjsai.39-5_E-O41
He YBian YDing XWu BGuan JZhang JZhou S(2024)Variate Associated Domain Adaptation for Unsupervised Multivariate Time Series Anomaly DetectionACM Transactions on Knowledge Discovery from Data10.1145/366357318:8(1-24)Online publication date: 3-May-2024
https://dl.acm.org/doi/10.1145/3663573
Show More Cited By

Index Terms

Multi-Source Deep Domain Adaptation with Weak Supervision for Time-Series Sensor Data
1. Computing methodologies
  1. Machine learning
    1. Learning paradigms
    2. Machine learning approaches
      1. Neural networks
2. Mathematics of computing
  1. Probability and statistics
    1. Statistical paradigms
      1. Time series analysis

Recommendations

ADATIME: A Benchmarking Suite for Domain Adaptation on Time Series Data
Unsupervised domain adaptation methods aim at generalizing well on unlabeled test data that may have a different (shifted) distribution from the training data. Such methods are typically developed on image data, and their application to time series data ...
Domain consistency regularization for unsupervised multi-source domain adaptive classification
Highlights
- We propose a novel multi-source domain adaptation method for classification.
- We ...
Abstract
Deep learning-based multi-source unsupervised domain adaptation (MUDA) has been actively studied in recent years. Compared with single-source unsupervised domain adaptation (SUDA), domain shift in MUDA exists not only between the ...
Mutual Domain Adaptation
Highlights
- We tackle a realistic problem setting of domain adaptation, where most domains are label-deficient and need to be helped and recent data become more sparsely labeled which makes the learning even more difficult.
- To tackle this problem, ...
Abstract
To solve the label sparsity problem, domain adaptation has been well-established, suggesting various methods such as finding a common feature space of different domains using projection matrices or neural networks. Despite recent advances, domain ...

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image ACM Conferences

KDD '20: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining

August 2020

3664 pages

ISBN:9781450379984

DOI:10.1145/3394486

General Chairs:
Rajesh Gupta
UC San Diego, USA
,
Yan Liu
USC, USA
,
Program Chairs:
Mohak Shah
LG Electronics, USA
,
Suju Rajan
Linkedin, USA
,
Publications Chairs:
Jiliang Tang
Michigan State, USA
,
B. Aditya Prakash
Georgia Tech, USA

Copyright © 2020 ACM.

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 20 August 2020

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article

Funding Sources

Conference

KDD '20

Sponsor:

KDD '20: The 26th ACM SIGKDD Conference on Knowledge Discovery and Data Mining

July 6 - 10, 2020

CA, Virtual Event, USA

Acceptance Rates

Overall Acceptance Rate 1,133 of 8,635 submissions, 13%

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

65
Total Citations
View Citations
3,059
Total Downloads

Downloads (Last 12 months)785
Downloads (Last 6 weeks)60

Reflects downloads up to 24 Sep 2024

Other Metrics

View Author Metrics

Citations

Cited By

Singh AChakraborty SHe ZPang YZhang SSubedi RLustria MCharness NBoot W(2024)Predicting Adherence to Computer-Based Cognitive Training Programs Among Older Adults: Study of Domain Adaptation and Deep LearningJMIR Aging10.2196/537937(e53793-e53793)Online publication date: 16-Sep-2024
https://doi.org/10.2196/53793
Oshima HIshizone THiguchi T(2024)Inter-seasons and Inter-households Domain Adaptation Based on DANNs and Pseudo Labeling for Non-Intrusive Occupancy Detection非侵入型在宅推定に対するDANNsと疑似ラベリングをもとにした季節間および世帯間の教師なしドメイン適応手法Transactions of the Japanese Society for Artificial Intelligence10.1527/tjsai.39-5_E-O4139:5(E-O41_1-13)Online publication date: 1-Sep-2024
https://doi.org/10.1527/tjsai.39-5_E-O41
He YBian YDing XWu BGuan JZhang JZhou S(2024)Variate Associated Domain Adaptation for Unsupervised Multivariate Time Series Anomaly DetectionACM Transactions on Knowledge Discovery from Data10.1145/366357318:8(1-24)Online publication date: 3-May-2024
https://dl.acm.org/doi/10.1145/3663573
Zhang JFeng LLiu ZWu YHe YDong YXu DBaeza-Yates RBonchi F(2024)Diverse Intra- and Inter-Domain Activity Style Fusion for Cross-Person Generalization in Activity RecognitionProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining10.1145/3637528.3671828(4213-4222)Online publication date: 25-Aug-2024
https://dl.acm.org/doi/10.1145/3637528.3671828
Wang JBai GCheng WChen ZZhao LChen HBaeza-Yates RBonchi F(2024)POND: Multi-Source Time Series Domain Adaptation with Information-Aware Prompt TuningProceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining10.1145/3637528.3671721(3140-3151)Online publication date: 25-Aug-2024
https://dl.acm.org/doi/10.1145/3637528.3671721
Lu WWang JSun XChen YJi XYang QXie X(2024)Diversify: A General Framework for Time Series Out-of-Distribution Detection and GeneralizationIEEE Transactions on Pattern Analysis and Machine Intelligence10.1109/TPAMI.2024.335521246:6(4534-4550)Online publication date: Jun-2024
https://doi.org/10.1109/TPAMI.2024.3355212
Ragab MEldele EChen ZWu MKwoh CLi X(2024)Self-Supervised Autoregressive Domain Adaptation for Time Series DataIEEE Transactions on Neural Networks and Learning Systems10.1109/TNNLS.2022.318325235:1(1341-1351)Online publication date: Jan-2024
https://doi.org/10.1109/TNNLS.2022.3183252
Ren LCheng X(2024)Single/Multi-Source Black-Box Domain Adaption for Sensor Time Series DataIEEE Transactions on Cybernetics10.1109/TCYB.2023.330083254:8(4712-4723)Online publication date: Aug-2024
https://doi.org/10.1109/TCYB.2023.3300832
Eldele ERagab MChen ZWu MKwoh CLi X(2024)Contrastive Domain Adaptation for Time-Series Via Temporal MixupIEEE Transactions on Artificial Intelligence10.1109/TAI.2023.32934735:3(1185-1194)Online publication date: Mar-2024
https://doi.org/10.1109/TAI.2023.3293473
Zhong XWang JFeng HTan R(2024)Rapid User-Adaptive Wearable Activity Recognition via Difference Decomposition2024 International Joint Conference on Neural Networks (IJCNN)10.1109/IJCNN60899.2024.10651473(1-7)Online publication date: 30-Jun-2024
https://doi.org/10.1109/IJCNN60899.2024.10651473
Show More Cited By

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Get this Publication

Media

Figures

Other

Tables

View Table of Contents