TinyML for Ubiquitous Edge AI
Open accessAbstract
TinyML is a fast-growing multidisciplinary field at the intersection of machine learning, hardware, and software, that focuses on enabling deep learning algorithms on embedded (microcontroller powered) devices operating at extremely low power range (mW range and below). TinyML addresses the challenges in designing power-efficient, compact deep neural network models, supporting software framework, and embedded hardware that will enable a wide range of customized, ubiquitous inference applications on battery-operated, resource-constrained devices. In this report, we discuss the major challenges and technological enablers that direct this field's expansion. TinyML will open the door to the new types of edge services and applications that do not rely on cloud processing but thrive on distributed edge inference and autonomous reasoning.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Pruning, quantization, and knowledge distillation shrink parameter counts and reduce runtime while preserving most predictive performance, though over-compression risks degrading behavior during rare or extreme events.
“Pruning, quantization, and knowledge distillation shrink parameter counts and reduce runtime while preserving most predictive performance, though over-compression risks degrading behavior during rare or extreme events [33,34].”
From Explainable AI and Multi-Agent Systems for Energy Management in IoT-Edge Environments: A State of the Art Review (2026) · cites this paper as [33]
Sections 3.1–3.1.4 describe pruning, quantization, and knowledge distillation as ways to reduce model size, with model compression aiming to preserve accuracy. The copy does not establish that all three reduce runtime or parameter counts: quantization reduces bits per weight. It also does not say that over-compression degrades performance during rare or extreme events. (LLM, conf=0.86)
Experimental — accuracy not yet measuredThe TinyML deployment pipeline involves quantization and pruning before deployment on microcontrollers to address power and memory limits.
“To address power and memory limits, models undergo quantization and pruning [33] before deployment on microcontrollers.”
From Explainable AI and Multi-Agent Systems for Energy Management in IoT-Edge Environments: A State of the Art Review (2026) · cites this paper as [33]
In real-time control loops, local inference avoids cloud-offload delays that would hinder responsiveness.
“In real-time control loops-such as adjusting HVAC output in response to occupancy-local inference avoids cloud-offload delays that would hinder responsiveness [33].”
From Explainable AI and Multi-Agent Systems for Energy Management in IoT-Edge Environments: A State of the Art Review (2026) · cites this paper as [33]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017) · arXiv · doi:10.48550/arxiv.1704.04861 · Cited by 19,015 · see citing works
Distilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204 · see citing works
ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices
Zhang, X., Zhou, X., Lin, M., & Sun, J. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00716 · Cited by 9,982 · see citing works
Federated Learning for Internet of Things: Recent Advances, Taxonomy, and Open Challenges
Khan, L. U., Saad, W., Han, Z., Hossain, E., & Hong, C. S. (2021) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2021.3090430 · Cited by 892 · see citing works
Benchmark analysis of representative deep neural network architectures
Bianco, S., Cadene, R., Celona, L., & Napoletano, P. (2018) · IEEE Access · doi:10.1109/access.2018.2877890 · Cited by 838 · see citing works
Convolutional neural networks for small-footprint keyword spotting
Sainath, T. N., & Parada, C. (2015) · Interspeech (USB) · doi:10.21437/interspeech.2015-352 · Cited by 529 · see citing works
TinyLSTMs: Efficient Neural Speech Enhancement for Hearing Aids
Fedorov, I., Stamenovic, M., Jensen, C., Yang, L.-C., Mandell, A., Gan, Y., Mattina, M., & Whatmough, P. N. (2020) · Interspeech 2020 · doi:10.21437/interspeech.2020-1864 · Cited by 116 · see citing works
Locally-connected and convolutional neural networks for small footprint speaker recognition
Chen, Y., Moreno, I. L., Sainath, T. N., Visontai, M., Álvarez, R., & Parada, C. (2015) · Interspeech (USB) · doi:10.21437/interspeech.2015-297 · Cited by 91 · see citing works
Yang, M., Yeh, C.-H., Zhou, Y., Cerqueira, J. P., Lazar, A. A., & Seok, M. (2019) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2019.2894360 · Cited by 82 · see citing works
On-device machine learning An algorithms and learning theory perspective
Dhar, S., Guo, J., Liu, J., Tripathi, S., Kurup, U., & Shah, M. (2019) · arXiv · doi:10.48550/arxiv.1911.00623 · Cited by 47 · see citing works
Faraone, A., & Delgado-Gonzalo, R. (2020) · 2020 2nd IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas48895.2020.9073950 · Cited by 36 · see citing works
Event-driven Pipeline for Low-latency Low-compute Keyword Spotting and Speaker Verification System
Ceolini, E., Anumula, J., Braun, S., & Liu, S. (2019) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2019.8683669 · Cited by 20 · see citing works
Flagged first, then by citations
1 of the 13 references couldn’t be matched to a known paper.
Cited by
Sorted by citations
Singh, R., & Gill, S. S. (2023) · Internet of Things and Cyber-Physical Systems · doi:10.1016/j.iotcps.2023.02.004 · Cited by 572 · see citing works
A Comprehensive Survey on TinyML
Abadade, Y., Temouden, A., Bamoumen, H., Benamar, N., Chtouki, Y., & Hafid, A. S. (2023) · IEEE Access · doi:10.1109/access.2023.3294111 · Cited by 366 · see citing works
“There are three methods for deploying models [62]: hand coding, code generation, and ML interpreters.”
MethodIntelligence at the Extreme Edge: A Survey on Reformable TinyML
Rajapakse, V., Karunanayake, I., & Ahmed, N. (2023) · ACM Computing Surveys · doi:10.1145/3583683 · Cited by 129 · see citing works
TinyML: A Systematic Review and Synthesis of Existing Research
Han, H., & Siebert, J. (2022) · International Conference on Artificial Intelligence in Information and Communication (ICAIIC) · doi:10.1109/icaiic54071.2022.9722636 · Cited by 117 · see citing works
A Review on the emerging technology of TinyML
Tsoukas, V., Gkogkidis, A., Boumpa, E., & Kakarountas, A. P. (2024) · ACM Computing Surveys · doi:10.1145/3661820 · Cited by 122 · see citing works
TinyML-Based Classification in an ECG Monitoring Embedded System
Kim, E., Kim, J., Park, J., Ko, H., & Kyung, Y. (2023) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2023.031663 · Cited by 98 · see citing works
Osman, A., Abid, U., Gemma, L., Perotto, M., & Brunelli, D. (2022) · Lecture notes in electrical engineering · doi:10.1007/978-3-030-95498-7_20 · Cited by 34 · see citing works
A Study on ECG Monitoring Embedded Systems
Kim, J., Kim, E., Kyung, Y., & Ko, H. (2022) · International Conference on Information and Communication Technology Convergence (ICTC) · doi:10.1109/ictc55196.2022.9952978 · Cited by 32 · see citing works
LOPdM: A Low-Power On-Device Predictive Maintenance System Based on Self-Powered Sensing and TinyML
Chen, Z., Gao, Y., & Liang, J. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3308251 · Cited by 33 · see citing works
Training-free neural architecture search: A review
Wu, M., & Tsai, C. (2023) · ICT Express · doi:10.1016/j.icte.2023.11.001 · Cited by 29 · see citing works
Quantized Transformer Language Model Implementations on Edge Devices
Rahman, M. W. U., Abrar, M. M., Copening, H. G., Hariri, S., Shao, S., Satam, P., & Salehi, S. (2023) · International Conference on Machine Learning and Applications · doi:10.1109/icmla58977.2023.00104 · Cited by 27 · see citing works
Aira, J., Olivares, T., Delicado, F. M., & Vezzani, D. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3265119 · Cited by 26 · see citing works
Álvarez-López, C., González-Briones, A., & Li, T. (2026) · Electronics · doi:10.3390/electronics15020385 · Cited by 23 · see citing works
Software engineering approaches for TinyML based IoT embedded vision
Lakshman, S. B., & Eisty, N. U. (2022) · doi:10.1145/3528227.3528569 · Cited by 18 · see citing works
How TinyML Can be Leveraged to Solve Environmental Problems: A Survey
Bamoumen, H., Temouden, A., Benamar, N., & Chtouki, Y. (2022) · International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT) · doi:10.1109/3ict56508.2022.9990661 · Cited by 16 · see citing works
Zhou, S., Guo, T., Luan, X., & Li, Y. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3405356 · Cited by 14 · see citing works
Dai, W., & Zhou, L. (2022) · International Conference on Pattern Recognition and Artificial Intelligence (PRAI) · doi:10.1109/prai55851.2022.9904109 · Cited by 9 · see citing works
DiTMoS: Delving into Diverse Tiny-Model Selection on Microcontrollers
Ma, X., He, S., Qiao, H., & Ma, D. (2024) · IEEE International Conference on Pervasive Computing and Communications · doi:10.1109/percom59722.2024.10494422 · Cited by 6 · see citing works
Schauer, J., Goodarzi, P., Schütze, A., & Schneider, T. (2025) · Journal of sensors and sensor systems · doi:10.5194/jsss-14-169-2025 · Cited by 5 · see citing works
ML Blocks: A Block-Based, Graphical User Interface for Creating TinyML Models
Williams, R., Moskal, M., & de Halleux, P. (2022) · IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC) · doi:10.1109/vl/hcc53370.2022.9833149 · Cited by 5 · see citing works
TinyAirNet: TinyML Model Transmission for Energy-Efficient Image Retrieval From IoT Devices
Shiraishi, J., Thorsager, M., Pandey, S. R., & Popovski, P. (2024) · IEEE Communications Letters · doi:10.1109/lcomm.2024.3436816 · Cited by 5 · see citing works
Explainable AI: A Way to Achieve Trustworthy AI
Li, Y., Xiao, Y., Gong, Y., Zhang, R., Huo, Y., & Wu, Y. (2024) · IEEE Conference on Big Data Security on Cloud (BigDataSecurity) · doi:10.1109/bigdatasecurity62737.2024.00034 · Cited by 4 · see citing works
Multiple Disease Detection using Machine Learning Techniques
Acharya, D., Eashwer, K., Kumar, S., Rajagopal, S., Kishoreraja, P. C., & Srinivasagan, R. (2023) · International Journal of Online and Biomedical Engineering (iJOE) · doi:10.3991/ijoe.v19i13.40523 · Cited by 4 · see citing works
An autonomous robot for removing Underwater pollution : OCTACLEANER
Reddy, A. N. N., Pavani, D., Rakshitha, B. S., & Saw, N. (2023) · IEEE International Conference on Technology, Engineering, Management for Societal Impact using Marketing, Entrepreneurship and Talent (TEMSMET) · doi:10.1109/temsmet56707.2023.10150012 · Cited by 3 · see citing works
Trees have Ears: An Acoustic Surveillance and TinyML-Based for Detecting Illegal Logging
Lorenzo, A. M., Barien, R., Favila, N. D., Basa, D., Ventura, J. M., & Catolos, S. (2024) · International Conference of Adisutjipto on Aerospace Electrical Engineering and Informatics (ICAAEEI) · doi:10.1109/icaaeei63658.2024.10899158 · Cited by 3 · see citing works
Privacy-Focused Edge AI-on-IoT with Federated Learning
Pal, S., Tan, W.-H., & Foo, Y.-L. (2023) · International Conference on Recent Advances and Innovations in Engineering · doi:10.1109/icraie59459.2023.10468159 · Cited by 3 · see citing works
Toward Diverse Tiny-Model Selection for Microcontrollers
Ma, X., He, S., Qiao, H., & Ma, D. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3561778 · Cited by 2 · see citing works
Enabling Efficient Deep Learning on MCU With Transient Redundancy Elimination
Liu, J., Zhang, F., Guan, J., Sung, H.-H., Guo, X., Long, S., Du, X., & Shen, X. (2024) · IEEE Transactions on Computers · doi:10.1109/tc.2024.3449102 · Cited by 2 · see citing works
Predictive Machine Maintenance Using Tiny ML
Bhide, A., Ghodake, D., Jamle, A., Shaikh, S., & Bhujbal, P. S. R. (2023) · International Journal for Research in Applied Science and Engineering Technology · doi:10.22214/ijraset.2023.51254 · Cited by 2 · see citing works
Practical Evaluation of Federated Learning in Edge AI for IoT
Pal, S., Umair, M., Tan, W.-H., & Foo, Y. (2023) · JOIV International Journal on Informatics Visualization · doi:10.30630/joiv.7.3-2.2329 · Cited by 1 · see citing works
Esmaeili, A., Ghorrati, Z., & Matson, E. T. (2023) · IEEE International Conference on Robotic Computing (IRC) · doi:10.1109/irc59093.2023.00068 · Cited by 1 · see citing works
Changes in Communication Networks and Computing for Intelligent IoT Services from Cloud to Edge
Kyung, Y., & Ko, H. (2022) · International Conference on Information and Communication Technology Convergence (ICTC) · doi:10.1109/ictc55196.2022.9952986
Real-Time Arabic Digit Spotting with TinyML-Optimized CNNs on Edge Devices
Adla, Y. A., Saghir, M. A. R., & Awad, M. (2023) · IFIP advances in information and communication technology · doi:10.1007/978-3-031-34111-3_44
Implementing TinyML-based Speech Recognition using Arduino
Koo, G.-S., & Seo, Y. G. (2023) · Journal of Digital Contents Society · doi:10.9728/dcs.2023.24.6.1285
Process Optimisation in Real-Time Domains Using Microcontroller-Based Sensor Devices
D'Souza, O. R. J., Mukhopadhyay, S. C., & Sheng, Q. Z. (2023) · International Conference on Sensing Technology · doi:10.1109/icst59744.2023.10460771
Ultralow power communication: Challenges and machine learning solutions
Sharma, R., Singh, B., Prashar, D., & Jha, S. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-29927-8.00012-6
- Authors
- Soro, S.
- Corresponding author
- Soro, Stanislava
- Venue
- arXiv
- Publisher
- arXiv
- Published
- 2021
- ISSN
- 2331-8422 (ISSN-L)
- arXiv
- 2102.01255
- OpenAlex
- W3126452686
- Cited by
- 36
- References
- 13
- Citations by year
- 2025: 4 · 2024: 8 · 2023: 16 · 2022: 8
- Type
- posted content
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Hardware and Architecture › Energy Efficiency in Computing
- Keywords
- TinyML, edge AI, embedded deep learning, microcontroller, low-power inference, resource-constrained devices, battery-powered devices, autonomous reasoning, edge services
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
