research-article

Open access

MLIoT: An End-to-End Machine Learning System for the Internet-of-Things

Authors:

Sudershan Boovaraghavan,

Prahaladha Mallela,

Yuvraj AgarwalAuthors Info & Claims

IoTDI '21: Proceedings of the International Conference on Internet-of-Things Design and Implementation

Pages 169 - 181

https://doi.org/10.1145/3450268.3453522

Published: 18 May 2021 Publication History

Abstract

Modern Internet of Things (IoT) applications, from contextual sensing to voice assistants, rely on ML-based training and serving systems using pre-trained models to render predictions. However, real-world IoT environments are diverse, with rich IoT sensors and need ML models to be personalized for each setting using relatively less training data. Most existing general-purpose ML systems are optimized for specific and dedicated hardware resources and do not adapt to changing resources and different IoT application requirements. To address this gap, we propose MLIoT, an end-to-end Machine Learning System tailored towards supporting the entire lifecycle of IoT applications. MLIoT adapts to different IoT data sources, IoT tasks, and compute resources by automatically training, optimizing, and serving models based on expressive application-specific policies. MLIoT also adapts to changes in IoT environments or compute resources by enabling re-training, and updating models served on the fly while maintaining accuracy and performance. Our evaluation across a set of benchmarks show that MLIoT can handle multiple IoT tasks, each with individual requirements, in a scalable manner while maintaining high accuracy and performance. We compare MLIoT with two state-of-the-art hand-tuned systems and a commercial ML system showing that MLIoT improves accuracy from 50% - 75% while reducing or maintaining latency.

References

[1]

Martín Abadi et al. 2016. TensorFlow: A System for Large-Scale Machine Learning. In Proc. of the 12th USENIX Conference on Operating Systems Design and Implementation (Savannah, GA, USA) (OSDI'16). USENIX Association, USA, 265--283.

[2]

Zeeshan Ahmed et al. 2019. Machine Learning at Microsoft with ML. NET. In Proc. of the 25th ACM SIGKDD Internat. Conference on Knowledge Discovery & Data Mining. ACM, New York, NY, USA, 2448--2458.

[3]

Amazon. 2020. What Is Alexa? https://developer.amazon.com/en-US/alexa.

[4]

Amazon AWS. 2020. Amazon Rekognition - Video and Image - AWS. https://aws.amazon.com/rekognition.

[5]

Amazon AWS. 2020. IoT Greengrass. https://aws.amazon.com/greengrass/.

[6]

Denis Baylor et al. 2017. TFX: A TensorFlow-Based Production-Scale Machine Learning Platform. In Proc. of the 23rd ACM SIGKDD Internat. Conference on KDD (Halifax, NS, Canada) (KDD '17). ACM, New York, NY, USA, 1387--1395. https://doi.org/10.1145/3097983.3098021

[7]

James Bergstra and Yoshua Bengio. 2012. Random search for hyper-parameter optimization. Journal of machine learning research 13, Feb (2012), 281--305.

Digital Library

[8]

Andreas Bulling, Jamie A Ward, Hans Gellersen, and Gerhard Troster. 2010. Eye movement analysis for activity recognition using electrooculography. IEEE transactions on pattern analysis and machine intelligence 33, 4 (2010), 741--753.

Digital Library

[9]

Cgroups. 2020. Control Groups --- The Linux Kernel. https://www.kernel.org/doc/html/latest/admin-guide/cgroup-v1/cgroups.html.

[10]

Carl Chalmers, Paul Fergus, et al. 2020. Detecting activities of daily living and routine behaviours in dementia patients living alone using smart meter load disaggregation. IEEE Transactions on Emerging Topics in Computing (2020).

[11]

Trishul Chilimbi et al. 2014. Project Adam: Building an Efficient and Scalable Deep Learning Training System. In Proc. of the 11th USENIX Conference on Operating Systems Design and Implementation (Broomfield, CO) (OSDI'14). USENIX Association, USA, 571--582.

[12]

Giorgio Conte et al. 2014. BlueSentinel: A First Approach Using IBeacon for an Energy Efficient Occupancy Detection System. In Proceedings of the 1st ACM Conference on Embedded Systems for Energy-Efficient Buildings (Memphis, Tennessee) (BuildSys '14). Association for Computing Machinery, New York, NY, USA, 11--19. https://doi.org/10.1145/2676061.2674078

[13]

Daniel Crankshaw et al. 2014. The Missing Piece in Complex Analytics: Low Latency, Scalable Model Management and Serving with Velox. CoRR abs/1409.3809 (2014). arXiv:1409.3809 http://arxiv.org/abs/1409.3809

[14]

Daniel Crankshaw et al. 2017. Clipper: A Low-Latency Online Prediction Serving System. In Proc. of the 14th USENIX Conf. on Networked Systems Design and Implementation (Boston, MA, USA) (NSDI'17). USENIX Association, USA, 613--627.

[15]

Daniel Crankshaw et al. 2018. InferLine: ML Inference Pipeline Composition Framework. CoRR abs (2018). arXiv:1812.01776 http://arxiv.org/abs/1812.01776

[16]

Everton de Matos et al. 2020. Context information sharing for the Internet of Things: A survey. Computer Networks 166 (2020), 106988.

Digital Library

[17]

Jeffrey Dean et al. 2012. Large scale distributed deep networks. In Advances in neural information processing systems. 1223--1231.

[18]

Christian Debes et al. 2016. Monitoring activities of daily living in smart homes: Understanding human behavior. IEEE Signal Process. Mag. 33, 2 (2016), 81--94.

[19]

Thomas G Dietterich. 2000. Ensemble methods in machine learning. In Internat. workshop on multiple classifier systems. Springer, 1--15.

[20]

Mohamed Faisal Elrawy, Ali Ismail Awad, and Hesham F. A. Hamed. 2018. Intrusion Detection Systems for IoT-Based Smart Environments: A Survey. 7, 1, Article 123 (Dec. 2018), 20 pages. https://doi.org/10.1186/s13677-018-0123-6

[21]

Antti J Eronen et al. 2005. Audio-based context recognition. IEEE Transactions on Audio, Speech, and Language Processing 14, 1 (2005), 321--329.

Digital Library

[22]

Anthony Fleury et al. 2010. SVM-Based Multimodal Classification of Activities of Daily Living in Health Smart Homes: Sensors, Algorithms, and First Experimental Results. Trans. Info. Tech. Biomed. 14, 2 (March 2010), 274--283. https://doi.org/10.1109/TITB.2009.2037317

[23]

Daniel Golovin et al. 2017. Google vizier: A service for black-box optimization. In Proc. of the 23rd ACM SIGKDD Internat. conference on KDD. 1487--1495.

[24]

Google. 2020. TensorFlow Core | Machine Learning for Beginners and Experts. https://www.tensorflow.org/overview.

[25]

Google Cloud. 2020. Cloud AutoML. https://cloud.google.com/automl.

[26]

Google Cloud. 2020. Derive Insights via ML. https://cloud.google.com/vision.

[27]

Google Nest. 2020. Google Nest Smart Speakers & Displays - Google Store. https://store.google.com/product/google_home.

[28]

Google Research. 2020. Coral: toolkit to build AI products. https://coral.ai/.

[29]

gRPC. 2020. gRPC-Overview. https://grpc.io/docs/.

[30]

Sidhant Gupta et al. 2010. ElectriSense: Single-Point Sensing Using EMI for Electrical Event Detection and Classification in the Home. In Proc. of the 12th ACM Internat. Conference on Ubiquitous Computing (Copenhagen, Denmark) (UbiComp '10). ACM, New York, NY, USA, 139--148. https://doi.org/10.1145/1864349.1864375

[31]

S. Hershey, S. Chaudhuri, D. P. W. Ellis, J. F. Gemmeke, A. Jansen, R. C. Moore, et al. 2017. CNN architectures for large-scale audio classification. In 2017 IEEE Internat. Conference on Acoustics, Speech and Signal Processing (ICASSP). 131--135. https://doi.org/10.1109/ICASSP.2017.7952132

Digital Library

[32]

Peter Hevesi et al. 2014. Monitoring household activities and user location with a cheap, unobtrusive thermal sensor array. In Proc. of the 2014 ACM Internat. joint conference on pervasive and ubiquitous computing. 141--145.

[33]

Intel. 2020. Intel® Neural Compute Stick: A Plug and Play Development Kit for AI Inferencing. https://software.intel.com/en-us/neural-compute-stick.

[34]

Jeffrey Dunn. 2020. FBLearner Flow: Facebook's AI backbone. https://engineering.fb.com/core-data/introducing-fblearner-flow-facebook-s-ai-backbone/.

[35]

Jeremy Hermann and others. 2020. Meet Michelangelo: Uber's Machine Learning Platform. https://eng.uber.com/michelangelo-machine-learning-platform/.

[36]

Ian T Jolliffe. 1986. Principal components in regression analysis. In Principal component analysis. Springer, 129--155.

[37]

Rushil Khurana et al. 2018. GymCam: Detecting, recognizing and tracking simultaneous exercises in unconstrained scenes. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 2, 4 (2018), 1--17.

Digital Library

[38]

Gierad Laput, Karan Ahuja, Mayank Goel, and Chris Harrison. 2018. Ubicoustics: Plug-and-Play Acoustic Activity Recognition. In Proc. of the 31st Annual ACM Symposium on UIST (Berlin, Germany) (UIST '18). ACM, New York, NY, USA, 213--224. https://doi.org/10.1145/3242587.3242609

Digital Library

[39]

Gierad Laput, Yang Zhang, and Chris Harrison. 2017. Synthetic Sensors: Towards General-Purpose Sensing. In Proc. of the 2017 CHI Conference on Human Factors in Computing Systems (Denver, Colorado, USA) (CHI '17). ACM, New York, NY, USA, 3986--3999. https://doi.org/10.1145/3025453.3025773

Digital Library

[40]

Yann LeCun and Corinna Cortes. 2010. MNIST handwritten digit database. http://yann.lecun.com/exdb/mnist/. (2010). http://yann.lecun.com/exdb/mnist/

[41]

Xin Lei, Andrew Senior, Alexander Gruenstein, and Jeffrey Sorensen. 2013. Accurate and compact large vocabulary speech recognition on mobile devices. (2013).

[42]

Mu Li et al. 2014. Scaling Distributed Machine Learning with the Parameter Server (OSDI '14). USENIX Association, USA, 583--598.

[43]

Laurens van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-SNE. Journal of machine learning research 9, Nov (2008), 2579--2605.

[44]

Mohammad Saeid Mahdavinejad et al. 2018. Machine learning for Internet of Things data analysis: A survey. Digital Communications and Networks 4, 3 (2018), 161--175.

[45]

P. McCullagh et al. 1989. Generalized Linear Models. Chapman & Hall, London.

[46]

Leland McInnes, John Healy, Nathaniel Saul, and Lukas Großberger. 2018. UMAP: Uniform Manifold Approximation and Projection. Journal of Open Source Software 3, 29 (2018). https://doi.org/10.21105/joss.00861

[47]

Microsoft. 2020. Azure IoT. https://azure.microsoft.com/en-us/overview/iot/.

[48]

Sebastian Mika et al. 1998. Kernel PCA and De-Noising in Feature Spaces (NIPS'98). MIT Press, Cambridge, MA, USA, 536--542.

[49]

Mites.io. 2020. Mites.io: a full-stack ubiquitous sensing platform. https://mites.io/.

[50]

Nvidia. 2020. Jetson Nano Developer Kit. https://developer.nvidia.com/embedded/jetson-nano-developer-kit.

[51]

Christopher Olston, Noah Fiedel, et al. 2017. Tensorflow-serving: Flexible, high-performance ml serving. arXiv preprint arXiv:1712.06139 (2017).

[52]

Adam Paszke et al. 2017. Automatic differentiation in PyTorch. (2017).

[53]

F. Pedregosa, G. Varoquaux, et al. 2011. Scikit-learn: Machine Learning in Python. Journal of Machine Learning Research 12 (2011), 2825--2830.

Digital Library

[54]

Raspberry Pi Foundation. 2020. Raspberry Pi 4 Model B. https://www.raspberrypi.org/products/raspberry-pi-4-model-b/.

[55]

Joseph Redmon et al. 2016. You only look once: Unified, real-time object detection. In Proc. of the IEEE conference on computer vision and pattern recognition. IEEE, Las Vegas, NV, USA, 779--788.

[56]

Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. 2015. Faster r-cnn: Towards real-time object detection with region proposal networks. In Advances in neural information processing systems. IEEE, Montreal, Quebec, Canada, 91--99.

Digital Library

[57]

Manaswi Saha et al. 2014. EnergyLens: Combining Smartphones with Electricity Meter for Accurate Activity Detection and User Annotation. In Proceedings of the 5th International Conference on Future Energy Systems (Cambridge, United Kingdom) (e-Energy '14). Association for Computing Machinery, New York, NY, USA, 289--300. https://doi.org/10.1145/2602044.2602058

[58]

Samsung. 2020. SmartThings. https://www.smartthings.com.

[59]

sandilands. 2020. Virtual Networking. https://sandilands.info/sgordon/virtnet.

[60]

Vin de Silva and Joshua B. Tenenbaum. 2002. Global versus Local Methods in Nonlinear Dimensionality Reduction. In Advances in neural information processing systems (NIPS'02). MIT Press, Cambridge, MA, USA, 721--728.

[61]

Jasper Snoek et al. 2012. Practical bayesian optimization of machine learning algorithms. In Advances in neural information processing systems. Curran Associates, Inc., Lake Tahoe, NV, 2951--2959.

[62]

M. Song, H. Li, and H. Wu. 2015. A Decentralized Load Balancing Architecture for Cache System. In 2015 Internat. Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery. IEEE, Xi'an, China, 114--119. https://doi.org/10.1109/CyberC.2015.44

[63]

Evan R Sparks, Shivaram Venkataraman, Tomer Kaftan, Michael J Franklin, and Benjamin Recht. 2017. Keystoneml: Optimizing pipelines for large-scale advanced analytics. In 2017 IEEE 33rd Internat. conference on data engineering (ICDE). IEEE, IEEE, New York, NY, USA, 535--546.

[64]

Edward J. Wang et al. 2015. MagnifiSense: Inferring Device Interaction Using Wrist-Worn Passive Magneto-Inductive Sensors. In Proc. of the 2015 ACM Internat. Joint Conference on Pervasive and Ubiquitous Computing(Osaka, Japan) (UbiComp '15). ACM, New York, NY, USA, 15--26. https://doi.org/10.1145/2750858.2804271

[65]

Wei Wang, Jinyang Gao, Meihui Zhang, et al. 2018. Rafiki: Machine Learning as an Analytics Service System. Proc. VLDB Endow. 12, 2 (Oct. 2018), 128--140. https://doi.org/10.14778/3282495.3282499

Digital Library

[66]

Doris Xin, Stephen Macke, Litian Ma, Jialin Liu, Shuchen Song, and Aditya Parameswaran. 2018. HELIX: Holistic Optimization for Accelerating Iterative Machine Learning. Proc. VLDB Endow. 12, 4 (Dec. 2018), 446--460. https://doi.org/10.14778/3297753.3297763

Digital Library

[67]

M. Xu, S. Alamro, T. Lan, and S. Subramaniam. 2017. LASER: A Deep Learning Approach for Speculative Execution and Replication of Deadline-Critical Jobs in Cloud. In 2017 26th Internat. Conference on Comp. Comm. and Networks (ICCCN). IEEE, Vancouver, BC, 1--8. https://doi.org/10.1109/ICCCN.2017.8038373

[68]

Matei Zaharia et al. 2012. Resilient Distributed Datasets: A Fault-Tolerant Abstraction for in-Memory Cluster Computing. In Proc. of the 9th USENIX Conference on Networked Systems Design and Implementation (San Jose, CA) (NSDI'12). USENIX Association, USA, 2.

[69]

Matei Zaharia et al. 2018. Accelerating the Machine Learning Lifecycle with MLflow. IEEE Data Eng. Bull. 41, 4 (2018), 39--45.

Cited By

Berraadi OGibet tani HBen ahmed M(2024)Improving Efficiency in IoT Data Streaming Through the Integration of Machine Learning: A ReviewProceedings of the 7th International Conference on Networking, Intelligent Systems and Security10.1145/3659677.3659679(1-8)Online publication date: 18-Apr-2024
https://dl.acm.org/doi/10.1145/3659677.3659679
Li DXu JYang ZMa QZhang LChen P(2024)LeoVR: Motion-Inspired Visual-LiDAR Fusion for Environment Depth EstimationIEEE Transactions on Mobile Computing10.1109/TMC.2023.333427123:6(7499-7516)Online publication date: Jun-2024
https://doi.org/10.1109/TMC.2023.3334271
Nguyen MTruong H(2024)On Optimizing Resources for Real‐Time End‐to‐End Machine Learning in Heterogeneous EdgesSoftware: Practice and Experience10.1002/spe.338355:3(541-558)Online publication date: 25-Oct-2024
https://doi.org/10.1002/spe.3383
Show More Cited By

Index Terms

MLIoT: An End-to-End Machine Learning System for the Internet-of-Things
1. Computer systems organization
  1. Architectures
    1. Distributed architectures
      1. Client-server architectures
2. Computing methodologies
  1. Artificial intelligence

Recommendations

ARES: Adaptive Resource-Aware Split Learning for Internet of Things
Abstract
Distributed training of Machine Learning models in edge Internet of Things (IoT) environments is challenging because of three main points. First, resource-constrained devices have large training times and limited energy budget. Second, ...
Optimization of physical education and training system based on machine learning and Internet of Things
Abstract
In order to improve the effect of physical education and training, this paper combines machine learning technology to identify sports training features and action prediction, and combines Internet of Things technology to process physical education ...
A Survey of Semi-Supervised Learning Methods
CIS '08: Proceedings of the 2008 International Conference on Computational Intelligence and Security - Volume 02

In traditional machine learning approaches to classification, one uses only a labelled set to train the classifier. Labelled instances however are often difficult, expensive, or time consuming to obtain, as they require the efforts of experienced human ...

Comments

Information & Contributors

Information

Published In

cover image ACM Conferences

IoTDI '21: Proceedings of the International Conference on Internet-of-Things Design and Implementation

May 2021

288 pages

ISBN:9781450383547

DOI:10.1145/3450268

Copyright © 2021 Owner/Author.

This work is licensed under a Creative Commons Attribution International 4.0 License.

Sponsors

Publisher

Association for Computing Machinery

New York, NY, United States

Publication History

Published: 18 May 2021

Permissions

Request permissions for this article.

Request Permissions

Check for updates

Author Tags

Qualifiers

Research-article
Research
Refereed limited

Conference

IoTDI '21

Sponsor:

SIGBED

IoTDI '21: International Conference on Internet-of-Things Design and Implementation

May 18 - 21, 2021

VA, Charlottesvle, USA

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

9
Total Citations
View Citations
904
Total Downloads

Downloads (Last 12 months)230
Downloads (Last 6 weeks)15

Reflects downloads up to 05 Mar 2025

Other Metrics

View Author Metrics

Citations

Cited By

Berraadi OGibet tani HBen ahmed M(2024)Improving Efficiency in IoT Data Streaming Through the Integration of Machine Learning: A ReviewProceedings of the 7th International Conference on Networking, Intelligent Systems and Security10.1145/3659677.3659679(1-8)Online publication date: 18-Apr-2024
https://dl.acm.org/doi/10.1145/3659677.3659679
Li DXu JYang ZMa QZhang LChen P(2024)LeoVR: Motion-Inspired Visual-LiDAR Fusion for Environment Depth EstimationIEEE Transactions on Mobile Computing10.1109/TMC.2023.333427123:6(7499-7516)Online publication date: Jun-2024
https://doi.org/10.1109/TMC.2023.3334271
Nguyen MTruong H(2024)On Optimizing Resources for Real‐Time End‐to‐End Machine Learning in Heterogeneous EdgesSoftware: Practice and Experience10.1002/spe.338355:3(541-558)Online publication date: 25-Oct-2024
https://doi.org/10.1002/spe.3383
Boovaraghavan SPatidar PAgarwal Y(2023)TAOProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies10.1145/36108967:3(1-32)Online publication date: 27-Sep-2023
https://dl.acm.org/doi/10.1145/3610896
Boovaraghavan SChen CMaravi ACzapik MZhang YHarrison CAgarwal Y(2023)MitesProceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies10.1145/35808657:1(1-32)Online publication date: 28-Mar-2023
https://dl.acm.org/doi/10.1145/3580865
Kim TPark HJin YLee SLee S(2023)Partition Placement and Resource Allocation for Multiple DNN-Based Applications in Heterogeneous IoT EnvironmentsIEEE Internet of Things Journal10.1109/JIOT.2023.323599310:11(9836-9848)Online publication date: 1-Jun-2023
https://doi.org/10.1109/JIOT.2023.3235993
Steidl MFelderer MRamler R(2023)The pipeline for the continuous development of artificial intelligence models—Current state of research and practiceJournal of Systems and Software10.1016/j.jss.2023.111615199:COnline publication date: 1-May-2023
https://dl.acm.org/doi/10.1016/j.jss.2023.111615
Dujić Rodić LStančić IZovko KPerković TŠolić P(2022)Tag Estimation Method for ALOHA RFID System Based on Machine Learning ClassifiersElectronics10.3390/electronics1116260511:16(2605)Online publication date: 19-Aug-2022
https://doi.org/10.3390/electronics11162605
Kayan HMajib YAlsafery WBarhamgi MPerera C(2021)AnoML-IoT: An end to end re-configurable multi-protocol anomaly detection pipeline for Internet of ThingsInternet of Things10.1016/j.iot.2021.10043716(100437)Online publication date: Dec-2021
https://doi.org/10.1016/j.iot.2021.100437

View Options

View options

PDF

View or Download as a PDF file.

eReader

View online with eReader.

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

Figures

Tables

Media

View Table of Conten