TinyML: Adopting tiny machine learning in smart cities
Abstract
Since Tiny machine learning (TinyML) is a quickly evolving subject, it is crucial that internet of things (IoT) devices be able to communicate with one another for the sake of stability and future development. TinyML is a rapidly growing subfield at the intersection of computer science, software engineering, and machine learning. Building deep learning (DL) networks that are a few hundred KBs in size has been the focus of recent research in this area. Deploying TinyML into small devices makes them smart. Reduced computation, power usage, and response time are just a few of the many advantages of TinyML. In this work, we provide an introduction to TinyML and demonstrate its benefits and architecture. Then, we investigate the meaning of quantization as a standard compression method for TinyML-related applications. There are two methods used to obtain the quantized weights of the deep learning models are quantization-aware training (QAT) and post-training quantization (PTQ), we described them in details. Next, TinyML-based solutions to improve the role of IoT devices in Smart Cities are highlighted as: lightweight training of deep learning models, inference of lightweight deep learning models in IoT devices, low power consumption, and inference of deep learning models in restricted resources of IoT devices. Finally, presenting some use cases for TinyML studies, with these studies applied to several cases in a variety of fields. To the best of the author’s knowledge, few studies have investigated TinyML as it is an emerging field.
Keywords
Full Text:
PDFReferences
1. Banbury CR, Reddi VJ, Lam M, et al. Benchmarking tinyml systems: Challenges and direction. arXiv 2020; arXiv:2003.04821. doi: 10.48550/arXiv.2003.04821
2. Lin J, Chen WM, Lin Y, et. al. Mcunet: Tiny deep learning on iot devices. Advances in Neural Information Processing Systems. 2020; 33: 11711-11722.
3. Warden P, Situnayake D. Tinyml: Machine learning with tensorflow lite on arduino and ultra-low-power microcontrollers. O’Reilly Media; 2019.
4. Ray PP. A review on TinyML: State-of-the-art and prospects. Journal of King Saud University-Computer and Information Sciences. 2022; 34(4): 1595-1623. doi: 10.1016/j.jksuci.2021.11.019
5. Estrebou CA, Fleming M, Saavedra MD, et al. Lightweight convolutional neural networks framework for really small tinyml devices. Communications in Computer and Information Science 2020; 1154: 29-39. doi: 10.1007/978-3-030-99170-8
6. Alajlan NN, Ibrahim DM. TinyML: Enabling of inference deep learning models on ultra-low-power IoT edge devices for AI applications. Micromachines. 2022; 13(6): 851. doi: 10.3390/mi13060851
7. TensorFlow Lite, TensorFlow, 2021. Available online: https://www.tensorflow.org/lite (accessed on 10 October 2023).
8. Dennis DK. Edgeml: machine learning for resource-constrained edge devices (2020). Available online: https://github.com/Microsoft/EdgeML (accessed on 10 October 2023).
9. Suda N, Loh D. Machine learning on arm cortex-m microcontrollers. Arm Ltd.: Cambridge, UK. 2019. Available online: https://www.arm.com/resources/guide/machine-learning-on-cortex-m (accessed on 10 October 2023).
10. X-CUBE-AI—AI Expansion Pack for STM32CubeMX—STMicroelectronics, 2021. Available online: https://www.st.com/en/embedded-software/x-cube-ai.html (accessed on 10 October 2023).
11. Sakr F, Bellotti F, Berta R, De Gloria A. Machine learning on mainstream microcontrollers. Sensors. 2020; 20(9): 2638. doi: 10.3390/s20092638
12. Sanchez-Iborra R, Skarmeta AF. Tinyml-enabled frugal smart objects: Challenges and opportunities. IEEE Circuits and Systems Magazine. 2020; 20(3): 4-18. doi: 10.1109/MCAS.2020.3005467
13. Puthal D, Mohanty SP, Wilson S, Choppali U. Collaborative edge computing for smart villages [energy and security]. IEEE Consumer Electronics Magazine. 2021; 10(3): 68-71. doi: 10.1109/MCE.2021.3051813
14. Raza W, Osman A, Ferrini F, Natale FD. Energy-efficient inference on the edge exploiting TinyML capabilities for UAVs. Drones. 2021; 5(4): 127. doi: 10.3390/drones
15. Huang Q. Weight-quantized squeezenet for resource-constrained robot vacuums for indoor obstacle classification. AI. 2022; 3(1): 180-193. doi: 10.3390/ai3010011
16. Model optimization TensorFlow Lite. Available online: https://www.tensorflow.org/lite/performance/model_optimization (accessed on 10 October 2023).
17. TensorFlow Model Optimization. Available online: https://www.tensorflow.org/model_optimization (accessed on 10 October 2023).
18. Novac PE, Boukli Hacene G, Pegatoquet A, et al. Quantization and deployment of deep neural networks on microcontrollers. Sensors. 2021; 21(9): 2984. doi: 10.3390/s21092984
19. Mohan P, Paul AJ, Chirania A. A tiny CNN architecture for medical face mask detection for resource-constrained endpoints. In: Innovations in Electrical and Electronic Engineering: Proceedings of ICEEE 2021 2021 May 25 (pp. 657-670). Springer Singapore. doi: 10.1007/978-981-16-0749-3_52
20. Introducing the Model Optimization Toolkit for Tensor Flow TensorFlow, 2018. Available online: https://medium.com/tensorflow/introducing-the-model-optimization-toolkit-for-tensorflow-254aca1ba0a3 (accessed on 10 October 2023).
21. Krishnamoorthi R. Quantizing deep convolutional networks for efficient inference: A whitepaper. arXiv 2018. arXiv:1806.08342.
22. Post Training Quantization. Available online: https://www.tensorflow.org/lite/performance/post_training_quantization (accessed on 10 October 2023).
23. Garifulla M, Shin J, Kim C, et al. A case study of quantizing convolutional neural networks for fast disease diagnosis on portable medical devices. Sensors. 2021; 22(1): 219. doi: 10.3390/s22010219
24. Signoretti G, Silva M, Andrade P, et al. An evolving tinyml compression algorithm for iot environments based on data eccentricity. Sensors. 2021; 21(12): 4153. doi: 10.3390/s21124153
25. Chen Y, Zheng B, Zhang Z, et al. Deep learning on mobile and embedded devices: State-of-the-art, challenges, and future directions. ACM Computing Surveys (CSUR). 2020; 53(4): 1-37. doi: 10.1145/3398209
26. Merenda M, Porcaro C, Iero D. Edge machine learning for ai-enabled iot devices: A review. Sensors. 2020; 20(9): 2533. doi: 10.3390/s20092533
27. Wang F, Zhang M, Wang X, et al. Deep learning for edge computing applications: A state-of-the-art survey. IEEE Access. 2020; 8: 58322-58336. doi: 10.1109/ACCESS.2020.2982411
28. Soro S. TinyML for ubiquitous edge AI. arXiv 2021. arXiv:2102.01255.
29. David R, Duke J, Jain A, et al. Tensorflow lite micro: Embedded machine learning for tinyml systems. Proceedings of Machine Learning and Systems. 2021; 3: 800-811.
30. Paul AJ, Mohan P, Sehgal S. Rethinking generalization in american sign language prediction for edge devices with extremely low memory footprint. In: 2020 IEEE Recent Advances in Intelligent Computational Systems (RAICS); 3 December 2020; pp. 147-152. IEEE.
31. de Prado M, Rusci M, Capotondi A, et al. Robustifying the deployment of tinyml models for autonomous mini-vehicles. Sensors. 2021; 21(4): 1339. doi: 10.3390/s21041339
32. Gorospe J, Mulero R, Arbelaitz O, et al. A generalization performance study using deep learning networks in embedded systems. Sensors. 2021; 21(4): 1031. doi: 10.3390/s21041031
33. Alajlan NN, Ibrahim DM. DDD. TinyML: A TinyML-Based Driver Drowsiness Detection Model Using Deep Learning. Sensors. 2023; 23(12): 5696. doi: 10.3390/s23125696
DOI: https://doi.org/10.32629/jai.v7i4.1186
Refbacks
- There are currently no refbacks.
Copyright (c) 2024 Norah N. Alajlan, Dina M. Ibrahim
License URL: https://creativecommons.org/licenses/by-nc/4.0/