MiniLearn: On-Device Learning for Low-Power IoT Devices
Abstract
Recent advances in machine learning enable new, intelligent applications in the Internet of Things. For example, today's smartwatches use Deep Neural Networks (DNNs) to detect and classify human activities. The training of DNNs, however, is done offline with previously collected and labeled datasets using extensive computational resources such as GPUs on cloud services. Once being quantized and deployed on an IoT device, a DNN commonly remains unchanged.
We argue that this static nature of trained DNNs strongly limits their flexibility to adapt to requirements that change dynamically. For example, the device may need to adjust on the fly to the limited memory and energy resources, but only the retraining or pruning of the DNN in the cloud can address these issues. Moreover, the user may need to add new classes or refine existing ones, due to different problem domains materializing dynamically. Retraining DNNs requires a high volume of data collected from IoT devices and transmitted to the cloud. However, IoT devices depend on energyefficient communication with limited reliability and network bandwidth. In addition, cloud storage of extensive IoT data raises significant privacy concerns.
This paper introduces MiniLearn that enables re-training of DNNs on resource-constrained IoT devices. MiniLearn allows IoT devices to re-train and optimize pre-trained, quantized neural networks using IoT data collected during deployment of an IoT device. We show that MiniLearn speeds up inference by a factor of up to 2 and requires up to 50% less memory compared to original DNN. In addition, MiniLearn increases classification accuracy for a sub-set by 3% to 9% of the original DNN.
References
[1]
Lin,S.,Ji,R.,Yan,C.,Zhang,B.,Cao,L.,Ye,Q.,Huang,F., and Doermann,D. 2019. "Towards optimal structured cnn pruning via generative adversarial learning". In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
[2]
David,R.,Duke,J.,Jain,A.,Reddi,V J.,Jeffries,N.,Li,J.,Kreeger,N.,Nappier,I.,Natraj,M.,Regev,S.,Rhodes,R.,Wang,T., and Warden,P. 2021. "Tensorflow lite micro: Embedded machine learning on tinyml systems". In Online-ArXiV Preprint or similar.
[3]
Kwapisz,J R.,Weiss,G M., and Moore,S A. 2010. "Activity recognition using cell phone accelerometers". In Proceedings of the Fourth International Workshop on Knowledge Discovery from Sensor Data. pp. 10--18.
[4]
Wu,Q.,Chen,X.,Zhou,Z., and Zhang,J. 2020. "Fedhome: Cloud-edge based personalized federated learning for in-home health monitoring". In IEEE Transactions on Mobile Computing. pp. 1--1.
[5]
Han,S.,Mao,H., and Dally,W J. 2016. "Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding". In International Conference on Learning Representations (ICLR).
[6]
Li,H.,Kadav,A.,Durdanovic,I.,Samet,H., and Graf,H P. 2016. "Pruning filters for efficient convnets". In Online-ArXiV Preprint or similar.
[7]
Chen,Y.,Qin,X.,Wang,J.,Yu,C., and Gao,W. 2020. "Fedhealth: A federated transfer learning framework for wearable healthcare". In IEEE Intelligent Systems. vol. 35,pp. 83--93.
[8]
Cai,H.,Chen,T.,Zhang,W.,Yu,Y., and Wang,J. 2018. "Efficient architecture search by network transformation". In Proceedings of the AAAI Conference on Artificial Intelligence. vol. 32,
[9]
Lai,L.,Suda,N., and Chandra,V. 2018. "Cmsis-nn: Efficient neural network kernels for arm cortex-m cpus". In Online-ArXiV Preprint or similar.
[10]
Warden,P. 2018. "Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition". In Online-ArXiV Preprint or similar.
[11]
Lin,J.,Chen,W.,Lin,Y.,Cohn,J.,Gan,C., and Han,S. 2020. "Mcunet: Tiny deep learning on iot devices". In Advances in Neural Information Processing Systems.
[12]
Pan,S J.,Kwok,J T., and Yang,Q. 2008. "Transfer learning via dimensionality reduction". In Proceedings of the AAAI Conference on Artificial Intelligence.
[13]
Chen,X.,Li,M.,Zhong,H.,Ma,Y., and Hsu,C.-H. 2022. "Dnnoff: Offloading dnn-based intelligent iot applications in mobile edge computing". In IEEE Transactions on Industrial Informatics. vol. 18,pp. 2820--2829.
[14]
Jacob,B.,Kligys,S.,Chen,B.,Zhu,M.,Tang,M.,Howard,A.,Adam,H., and Kalenichenko,D. 2018. "Quantization and training of neural networks for efficient integer-arithmetic-only inference". In IEEE Conference on Computer Vision and Pattern Recognition, CVPR '18.
[15]
Leem,S.-G.,Yoo,I.-C., and Yook,D. 2019. "Multitask learning of deep neural network-based keyword spotting for iot devices". In IEEE Transactions on Consumer Electronics. vol. 65,pp. 188--194.
[16]
Lee,S.,Nirjon,S. 2020. "Fast and scalable in-memory deep multitask learning via neural weight virtualization". In ACM International Conference on Mobile Systems, Applications, and Services, MobiSys '20.
[17]
Pham,H.,Guan,M.,Zoph,B.,Le,Q., and Dean,J. 2018. "Efficient neural architecture search via parameters sharing". In Proceedings of the 35th International Conference on Machine Learning. PMLR. pp. 10--15.
[18]
Bai,J.,Yuan,A.,Xiao,Z.,Zhou,H.,Wang,D.,Jiang,H., and Jiao,L. 2020. "Class incremental learning with few-shots based on linear programming for hyperspectral image classification". In IEEE Transactions on Cybernetics.
[19]
Zhao,R.,Hu,Y.,Dotzel,J.,Sa,C De., and Zhang,Z. 2019. "Improving neural network quantization without retraining using outlier channel splitting". In International Conference on Machine Learning, ICML '17.
[20]
Chandran,N.,Gupta,D.,Rastogi,A.,Sharma,R., and Tripathi,S. 2019. "Ezpc: Programmable and efficient secure two-party computation for machine learning". In 2019 IEEE European Symposium on Security and Privacy (EuroS P). pp. 496--511.
[21]
Profentzas,C.,Almgren,M., and Landsiedel,O. 2021. "Performance of deep neural networks on low-power iot devices". In ACM CPS-IoTBench '21.
[22]
Ren,J.,Wang,H.,Hou,T.,Zheng,S., and Tang,C. 2019. "Federated learning-based computation offloading optimization in edge computing-supported internet of things". In IEEE Access. vol. 7,pp. 69194--69201.
[23]
Sung,F.,Yang,Y.,Zhang,L.,Xiang,T.,Torr,P H., and Hospedales,T M. 2018. "Learning to compare: Relation network for few-shot learning". In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
[24]
Cai,Z.,He,X.,Sun,J., and Vasconcelos,N. 2017. "Deep learning with low precision by half-wave gaussian quantization". In IEEE Conference on Computer Vision and Pattern Recognition.
[25]
Ren,H.,Anicic,D., and Runkler,T. 2021. "Tinyol: Tinyml with onlinelearning on microcontrollers". In Online-ArXiV Preprint or similar.
[26]
Gouda,S K.,Kanetkar,S.,Harrison,D., and Warmuth,M K. 2020. "Speech recognition: Keyword spotting through image recognition". In Online-ArXiV Preprint or similar.
[27]
Chung,H.,Iorga,M.,Voas,J., and Lee,S. 2017. "alexa, can i trust you?". In IEEE Computer Journal. vol. 50,pp. 100--104.
[28]
Dong,Z.,Yao,Z.,Gholami,A.,Mahoney,M W., and Keutzer,K. 2019. "Hawq: Hessian aware quantization of neural networks with mixed-precision". In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV).
[29]
Graepel,T.,Lauter,K., and Naehrig,M. 2013. "Ml confidential: Machine learning on encrypted data". In Information Security and Cryptology. pp. 1--21.
[30]
Lee,S.-M.,Yoon,S M., and Cho,H. 2017. "Human activity recognition from accelerometer data using convolutional neural network". In IEEE International Conference on Big Data and Smart Computing, BigComp '17.
[31]
Eshratifar,A E.,Pedram,M. 2018. "Energy and performance efficient computation offloading for deep neural networks in a mobile cloud computing environment". In Proceedings of the 2018 on Great Lakes Symposium on VLSI, GLSVLSI '18. pp. 111--116.
Recommendations
Towards a green and secure architecture for reconfigurable IoT end-devices
ICCPS '18: Proceedings of the 9th ACM/IEEE International Conference on Cyber-Physical SystemsWith the advent of the Internet of Things (IoT), objects are becoming smaller, smarter and increasingly connected. IoT devices are being deployed in massive numbers, and the success of this new Internet era is heavily dependent upon the trust and ...
Robustness Analysis of Deep Learning Frameworks on Mobile Platforms
Testing Software and SystemsAbstractWith the recent increase in the computational power of modern mobile devices, machine learning-based heavy tasks such as face detection and speech recognition are now integral parts of such devices. This requires frameworks to execute machine ...
On Track of Sigfox Confidentiality with End-to-End Encryption
ARES '18: Proceedings of the 13th International Conference on Availability, Reliability and SecurityThe last years brought many novel challenges for the Internet of Things (IoT). Low capital and operational expenditures, massive deployments of devices, reliability and security are among the most crucial ones. The recently introduced Low-power wide area ...
Comments
Information & Contributors
Information
Published In
December 2022
273 pages
Sponsors
In-Cooperation
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 18 January 2023
Check for updates
Author Tags
Qualifiers
- Article
Conference
EWSN '22
Sponsor:
Acceptance Rates
EWSN '22 Paper Acceptance Rate 18 of 46 submissions, 39%;
Overall Acceptance Rate 81 of 195 submissions, 42%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 13 Jan 2025