research-article

Public Access

DeActive: Scaling Activity Recognition with Active Deep Learning

Authors:

H. M. Sajjad Hossain,

MD Abdullah Al Haiz Khan,

Nirmalya RoyAuthors Info & Claims

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, Volume 2, Issue 2

Article No.: 66, Pages 1 - 23

https://doi.org/10.1145/3214269

Published: 05 July 2018 Publication History

Abstract

Deep learning architectures have been applied increasingly in multi-modal problems which has empowered a large number of application domains needing much less human supervision in the process. As unlabeled data are abundant in most of the application domains, deep architectures are getting increasingly popular to extract meaningful information out of these large volume of data. One of the major caveat of these architectures is that the training phase demands both computational time and system resources much higher than shallow learning algorithms and it is posing a difficult challenge for the researchers to implement the architectures in low-power resource constrained devices. In this paper, we propose a deep and active learning enabled activity recognition model, DeActive, which is optimized according to our problem domain and reduce the resource requirements. We incorporate active learning in the process to minimize the human supervision along with the effort needed for compiling ground truth. The DeActive model has been validated using real data traces from a retirement community center (IRB #HP-00064387) and 4 public datasets. Our experimental results show that our model can contribute better accuracy while ensuring less amount of resource usages in reduced time compared to other traditional deep learning approaches in activity recognition.

References

[1]

2010. Wireless Sensor Tags. http://wirelesstag.net.

[2]

2015. UMBC-UMB Partnership Awards a Catalyst for Collaboration. http://research.umbc.edu/umbc-research-news/?id=49901.

[3]

2017. Torch: A scientific computing framework for LuaJIT. http://torch.ch/.

[4]

Hande Alemdar, TLM van Kasteren, and Cem Ersoy. 2017. Active learning with uncertainty sampling for large scale activity recognition in smart homes. Journal of Ambient Intelligence and Smart Environments 9, 2 (2017), 209--223.

[5]

Moez Baccouche and Franck Mamalet. {n. d.}. Sequential Deep Learning for Human Action Recognition. In Proc. of IEEE HBU, 2011.

Digital Library

[6]

Marc Bächlin and et al. 2010. Wearable assistant for Parkinson's disease patients with the freezing of gait symptom. IEEE Trans. Information Technology in Biomedicine (2010).

Digital Library

[7]

Ling Bao and Stephen S. Intille. 2004. Activity Recognition from User-Annotated Acceleration Data. In Proc.of PERVASIVE.

[8]

Sourav Bhattacharya and Nicholas D. Lane. 2016. Sparsification and Separation of Deep Learning Layers for Constrained Resource Inference on Wearables. In Proceedings of the 14th ACM Conference on Embedded Network Sensor Systems CD-ROM (SenSys '16). 176--189.

Digital Library

[9]

Sourav et al. Bhattacharya. 2014. Using Unlabeled Data in a Sparse-coding Framework for Human Activity Recognition. Pervasive Mob. Computing. (2014).

Digital Library

[10]

Monica Bianchini and Franco Scarselli. 2014. On the Complexity of Neural Network Classifiers: A Comparison Between Shallow and Deep Architectures. IEEE Trans. Neural Netw. Learning Syst. (2014).

[11]

Leon Bottou and Yoshua Bengio. 1995. Convergence properties of the k-means algorithms. In Advances in neural information processing systems. 585--592.

Digital Library

[12]

Yu chen ho and Ching hu lu. 2009. Active-learning assisted self-reconfigurable activity recognition in a dynamic environment. In ICRA.

[13]

Diane J Cook and Maureen Schmitter-Edgecombe. 2009. Assessing the quality of activities in a smart environment. Methods of information in medicine (2009).

[14]

M. Cornacchia, K. Ozcan, Y. Zheng, and S. Velipasalar. 2017. A Survey on Activity Detection and Classification Using Wearable Sensors. IEEE Sensors Journal 17, 2 (2017), 386--403.

[15]

Sanjoy Dasgupta. 2011. Two faces of active learning. Theor. Comput. Sci. (2011).

Digital Library

[16]

Tom Diethe, Niall Twomey, and Peter Flach. 2015. Bayesian Active Transfer Learning in Smart Homes. In Proc. of ICML.

[17]

Tushar Dobhal and Vivswan Shitole. 2015. Human Activity Recognition using Binary Motion Image and Deep Learning. Procedia Computer Science (2015).

[18]

Charles Elkan. 2003. Using the triangle inequality to accelerate k-means. In Proc. of ICML.

Digital Library

[19]

Andreas Bulling et al. 2014. A tutorial on human activity recognition using body-worn inertial sensors. ACM Comput. Surv. (2014).

Digital Library

[20]

Adam Coates et al. {n. d.}. Learning Feature Representations with K-Means. In Neural Networks: Tricks of the Trade - Second Edition, 2012.

[21]

Francisco Javier et al. 2016. Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition. Sensors (2016).

[22]

Gina Sprint et al. 2016. Unsupervised detection and analysis of changes in everyday physical activity data. Journal of Biomedical Informatics (2016).

Digital Library

[23]

H. M. Sajjad Hossain et al. 2016. Active learning enabled activity recognition. In IEEE PerCom.

[24]

Mohammad Abu Alsheikh et al. 2016. Deep Activity Recognition Models with Triaxial Accelerometers. AAAI Workshop (2016).

[25]

Michalis Vrigkas et al. 2015. A Review of Human Activity Recognition Methods. Front. Robotics and AI (2015).

[26]

Ming Zeng et al. 2014. Convolutional Neural Networks for human activity recognition using mobile sensors. In MobiCASE.

[27]

Raj Kumar Gupta et al. 2013. Human activities recognition using depth images. In Proc. of ACM Multimedia Conference.

Digital Library

[28]

Sourav Bhattacharya et al. {n. d.}. From smart to deep: Robust activity recognition on smartwatches using deep learning. In Proc. of PerCom Workshops 2016.

[29]

Scott E. Reed et al. 2014. Training Deep Neural Networks on Noisy Labels with Bootstrapping. CoRR (2014).

[30]

Tianqi Chen et al. 2016. Training Deep Nets with Sublinear Memory Cost. CoRR (2016).

[31]

Tâm Huynh et al. 2008. Discovery of activity patterns using topic models. In Proc. of UbiComp.

Digital Library

[32]

Kyle D. Feuz and Diane J. Cook. 2015. Transfer Learning Across Feature-Rich Heterogeneous Feature Spaces via Feature-Space Remapping (FSR). ACM Trans. Intell. Syst. Technol. (2015).

Digital Library

[33]

Yu Guan and Thomas Ploetz. 2017. Ensembles of Deep LSTM Learners for Activity Recognition using Wearables. arXiv preprint arXiv:1703.09370 (2017).

Digital Library

[34]

Nils Y. Hammerla and Shane Halloran. {n. d.}. Deep, Convolutional, and Recurrent Models for Human Activity Recognition Using Wearables. In Proc. of IJCAI 2016.

Digital Library

[35]

Jonathan Ho and Stefano Ermon. 2016. Generative Adversarial Imitation Learning. CoRR abs/1606.03476 (2016).

[36]

Yu-chen Ho, Ching-hu Lu, I-han Chen, Shih-shinh Huang, Ching-yao Wang, and Li-chen Fu. 2009. Active-learning Assisted Self-reconigurable Activity Recognition in a Dynamic Environment. In Proceedings of the 2009 IEEE International Conference on Robotics and Automation (ICRA'09). IEEE Press, Piscataway, NJ, USA, 1567--1572. http://dl.acm.org/citation.cfm?id=1703435.1703688

Digital Library

[37]

E. Hoque and J. Stankovic. 2012. AALO: Activity recognition in smart homes using Active Learning in the presence of Overlapped activities. In Proc. of PervasiveHealth.

[38]

H M. Sajjad Hossain, Nirmalya Roy, and MD Abdullah Al Hafiz Khan. 2015. Sleep Well: A Sound Sleep Monitoring Framework for Community Scaling. In Proceedings of the 2015 16th IEEE International Conference on Mobile Data Management - Volume 01 (MDM '15). 44--53.

Digital Library

[39]

Tâm Huynh and Bernt Schiele. 2006. Unsupervised Discovery of Structure in Activity Data Using Multiple Eigenspaces. In Proc. of LoCA.

Digital Library

[40]

Wenchao Jiang and Zhaozheng Yin. {n. d.}. Human Activity Recognition Using Wearable Sensors by Deep Convolutional Neural Networks. In Proc. of ACM MM 2015.

Digital Library

[41]

H M Sajjad Hossain et al Joseph Taylor. 2017. SenseBox: A Low-Cost Smart Home System. In PerCom Demo Workshop.

[42]

Tapas et al. Kanungo. 2002. An efficient k-means clustering algorithm: Analysis and implementation. IEEE transactions on pattern analysis and machine intelligence (2002).

Digital Library

[43]

Md Abdullah Al Hafiz Khan and Ruthvik Kukkapalli. 2016. RAM: Radar-based activity monitor. In Proc. of IEEE INFOCOM.

[44]

Narayanan Chatapuram Krishnan and Diane J. Cook. 2014. Activity recognition on streaming sensor data. Pervasive and Mobile Computing (2014).

Digital Library

[45]

Jennifer R. Kwapisz and et al. 2010. Activity recognition using cell phone accelerometers. SIGKDD Explorations (2010).

Digital Library

[46]

N. D. Lane and S. Bhattacharya. 2016. DeepX: A Software Accelerator for Low-Power Deep Learning Inference on Mobile Devices. In Proc. of IPSN.

Digital Library

[47]

Nicholas D. Lane and Petko Georgiev. {n. d.}. Can Deep Learning Revolutionize Mobile Sensing?. In Proc. of HotMobile '15.

Digital Library

[48]

Nicholas D. Lane and Petko Georgiev. {n. d.}. DeepEar: robust smartphone audio sensing in unconstrained acoustic environments using deep learning. In Proc. of ACM UbiComp 2015.

Digital Library

[49]

Walter S. Lasecki and Young Chol Song. 2013. Real-time Crowd Labeling for Deployable Activity Recognition. In Proc. of CSCW.

Digital Library

[50]

Honglak Lee and Alexis Battle. {n. d.}. Efficient sparse coding algorithms. In Advances in neural information processing systems, 2016.

Digital Library

[51]

Xinyu Li, Dongyang Yao, and et al. 2016. Activity recognition for medical teamwork based on passive RFID. In 2016 IEEE International Conference on RFID, RFID 2016. 34--42.

[52]

Xinyu et al. Li. {n. d.}. Deep Learning for RFID-Based Activity Recognition (SenSys '16).

[53]

Rong Liu, Ting Chen, and Lu Huang. 2010. Research on human activity recognition based on active learning. In Proc. of ICMLC.

[54]

Yonggang Lu, Ye Wei, Li Liu, Jun Zhong, Letian Sun, and Ye Liu. 2017. Towards unsupervised physical activity recognition using smartphone accelerometers. Multimedia Tools and Applications 76, 8 (2017), 10701--10719.

Digital Library

[55]

Juarez Monteiro, Roger Granada, Rodrigo C Barros, and Felipe Meneguzzi. 2017. Deep neural networks for kitchen activity recognition. In Neural Networks (IJCNN), 2017 International Joint Conference on. IEEE, 2048--2055.

[56]

Francisco Javier Ordóñez Morales and Daniel Roggen. 2016. Deep Convolutional Feature Transfer Across Mobile Activity Recognition Domains, Sensor Modalities and Locations. In Proceedings of the 2016 ACM International Symposium on Wearable Computers (ISWC '16). 92--99.

Digital Library

[57]

Youngja Park and et al. 2014. Generating balanced classifier-independent training samples from unlabeled data. Knowl. Inf. Syst. (2014).

Digital Library

[58]

Thomas Plötz, Nils Y. Hammerla, and Patrick Olivier. 2011. Feature Learning for Activity Recognition in Ubiquitous Computing. In Proc. of IJCAI 2011.

Digital Library

[59]

David Martin Ward Powers. 2011. Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation. International Journal of Machine Learning Technology (2011).

[60]

Valentin Radu and Nicholas D. Lane. {n. d.}. Towards Multimodal Deep Learning for Activity Recognition on Mobile Devices. In Proc. of UbiComp '16.

Digital Library

[61]

Daniel Roggen and Alberto Calatroni. 2010. Collecting complex activity datasets in highly rich networked sensor environments. In Proc. of INSS.

[62]

Paul E. Rybski and Manuela M. Veloso. 2005. Robust Real-Time Human Activity Recognition from Tracked Face Displacements. In Proc. of EPIA, 2005.

Digital Library

[63]

D. Sculley. 2010. Web-scale K-means Clustering. In Proceedings of the 19th International Conference on World Wide Web (WWW). 1177--1178.

Digital Library

[64]

Fabian Stark and Rudolph Triebel. 2015. CAPTCHA Recognition with Active Deep Learning.

[65]

M. et al. Stikic. 2008. Exploring semi-supervised and active learning for activity recognition. In Proc. of ISWC.

Digital Library

[66]

Cheng Tang and Claire Monteleoni. 2017. Convergence Rate of Stochastic k-means. In Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, AISTATS 2017, 20-22 April 2017, Fort Lauderdale, FL, USA. 1495--1503.

[67]

Dan Wang and Yi Shang. {n. d.}. A new active labeling method for deep learning. In Proc. of IJCNN 2014.

[68]

Keze Wang, Dongyu Zhang, Ya Li, Ruimao Zhang, and Liang Lin. 2017. Cost-Effective Active Learning for Deep Image Classification. CoRR abs/1701.03551 (2017).

Digital Library

[69]

Junyuan Xie, Ross B. Girshick, and Ali Farhadi. 2015. Unsupervised Deep Embedding for Clustering Analysis. CoRR (2015).

[70]

Shusen Zhou and Qingcai Chen. {n. d.}. Active Deep Networks for Semi-Supervised Sentiment Classification. In Proc. of COLING.

Digital Library

Cited By

Kianpisheh MMariakakis ATruong K(2024)exHARProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies10.1145/36435008:1(1-30)Online publication date: 6-Mar-2024
https://dl.acm.org/doi/10.1145/3643500
Zhang XCui SZhu TChen LZhou FNing H(2024)CASL: Capturing Activity Semantics Through Location Information for Enhanced Activity RecognitionIEEE/ACM Transactions on Computational Biology and Bioinformatics10.1109/TCBB.2023.323806421:4(1051-1059)Online publication date: Jul-2024
https://doi.org/10.1109/TCBB.2023.3238064
Wei BYi CZhang QZhu HZhu JJiang F(2024)ActiveSelfHAR: Incorporating Self-Training Into Active Learning to Improve Cross-Subject Human Activity RecognitionIEEE Internet of Things Journal10.1109/JIOT.2023.331415011:4(6833-6847)Online publication date: 15-Feb-2024
https://doi.org/10.1109/JIOT.2023.3314150
Show More Cited By

Index Terms

DeActive: Scaling Activity Recognition with Active Deep Learning

Recommendations

NuActiv: recognizing unseen new activities using semantic attribute-based learning
MobiSys '13: Proceeding of the 11th annual international conference on Mobile systems, applications, and services

We study the problem of how to recognize a new human activity when we have never seen any training example of that activity before. Recognizing human activities is an essential element for user-centric and context-aware applications. Previous studies ...
Deep appearance and motion learning for egocentric activity recognition

Egocentric activity recognition has recently generated great popularity in computer vision due to its widespread applications in egocentric video analysis. However, it poses new challenges comparing to the conventional third-person activity recognition ...
Smartwatch based activity recognition using active learning
CHASE '17: Proceedings of the Second IEEE/ACM International Conference on Connected Health: Applications, Systems and Engineering Technologies

Human activity monitoring has become widely popular in recent years, and has been utilized in a vast number of fields and applications. Most of the activity recognition algorithms proposed have emphasized the use of inertial sensors in smartphone ...

Comments

Information & Contributors

Information

Published In

cover image Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies Volume 2, Issue 2

June 2018

741 pages

EISSN:2474-9567

DOI:10.1145/3236498

Issue’s Table of Contents

Copyright © 2018 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]

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 05 July 2018

Accepted: 01 April 2018

Revised: 01 February 2018

Received: 01 August 2017

Published in IMWUT Volume 2, Issue 2

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article
Research
Refereed

Funding Sources

Alzheimer's Association
Office of Naval Research

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

51
Total Citations
View Citations
1,108
Total Downloads

Downloads (Last 12 months)135
Downloads (Last 6 weeks)25

Reflects downloads up to 12 Sep 2024

Other Metrics

View Author Metrics

Citations

Cited By

Kianpisheh MMariakakis ATruong K(2024)exHARProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies10.1145/36435008:1(1-30)Online publication date: 6-Mar-2024
https://dl.acm.org/doi/10.1145/3643500
Zhang XCui SZhu TChen LZhou FNing H(2024)CASL: Capturing Activity Semantics Through Location Information for Enhanced Activity RecognitionIEEE/ACM Transactions on Computational Biology and Bioinformatics10.1109/TCBB.2023.323806421:4(1051-1059)Online publication date: Jul-2024
https://doi.org/10.1109/TCBB.2023.3238064
Wei BYi CZhang QZhu HZhu JJiang F(2024)ActiveSelfHAR: Incorporating Self-Training Into Active Learning to Improve Cross-Subject Human Activity RecognitionIEEE Internet of Things Journal10.1109/JIOT.2023.331415011:4(6833-6847)Online publication date: 15-Feb-2024
https://doi.org/10.1109/JIOT.2023.3314150
Wu YMin FChen GShen SWen ZZhou X(2024)Self-supervised class-balanced active learning with uncertainty-mastery fusionKnowledge-Based Systems10.1016/j.knosys.2024.112192300(112192)Online publication date: Sep-2024
https://doi.org/10.1016/j.knosys.2024.112192
Park HLee GHan JChoi J(2024)Multiclass autoencoder-based active learning for sensor-based human activity recognitionFuture Generation Computer Systems10.1016/j.future.2023.09.029151(71-84)Online publication date: Feb-2024
https://doi.org/10.1016/j.future.2023.09.029
Smith RDragone M(2023)Generalisable Dialogue-based Approach for Active Learning of Activities of Daily LivingACM Transactions on Interactive Intelligent Systems10.1145/361601713:3(1-37)Online publication date: 11-Sep-2023
https://dl.acm.org/doi/10.1145/3616017
Wang WLi Q(2023)Generalized Zero-Shot Activity Recognition with Embedding-Based MethodACM Transactions on Sensor Networks10.1145/358269019:3(1-25)Online publication date: 5-Apr-2023
https://dl.acm.org/doi/10.1145/3582690
Cao YDhekne AAmmar M(2023)ViSigProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies10.1145/35807977:1(1-27)Online publication date: 28-Mar-2023
https://dl.acm.org/doi/10.1145/3580797
Liberis ELane N(2023)Differentiable Neural Network Pruning to Enable Smart Applications on MicrocontrollersProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies10.1145/35694686:4(1-19)Online publication date: 11-Jan-2023
https://dl.acm.org/doi/10.1145/3569468
Li QWang J(2023)FedAPI: Privacy-preserving Multi-end Adaptive Personal Identification via Federated Learning2023 International Conference on Artificial Intelligence of Things and Systems (AIoTSys)10.1109/AIoTSys58602.2023.00051(198-205)Online publication date: 19-Oct-2023
https://doi.org/10.1109/AIoTSys58602.2023.00051
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 Article

Media

Figures

Other

Tables

View Issue’s Table of Contents