Are Existing Knowledge Transfer Techniques Effective For Deep Learning on Edge Devices?
The DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
With the emergence of edge computing paradigm, many applications such as image recognition and augmented reality require to perform machine learning (ML) and artificial intelligence (AI) tasks on edge devices. Most AI and ML models are large and computational-heavy, whereas edge devices are usually equipped with limited computational and storage resources. Such models can be compressed and reduced for deployment on edge devices, but they may loose their capability and not perform well. Recent works used knowledge transfer techniques to transfer information from a large network (termed teacher) to a small one (termed student) in order to improve the performance of the latter. This approach seems to be promising for learning on edge devices, but a thorough investigation on its effectiveness is lacking. This paper provides an extensive study on the performance (in both accuracy and convergence speed) of knowledge transfer, considering different student architectures and different techniques for transferring knowledge from teacher to student. The results show that the performance of KT does vary by architectures and transfer techniques. A good performance improvement is obtained by transferring knowledge from both the intermediate layers and last layer of the teacher to a shallower student. But other architectures and transfer techniques do not fare so well and some of them even lead to negative performance impact.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
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
TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
Abadi, M., Agarwal, A., Barham, P., Brevdo, E., Chen, Z., Citro, C., Corrado, G. S., Davis, A., Dean, J., Devin, M., et al. (2016) · arXiv · doi:10.48550/arxiv.1603.04467 · Cited by 9,408
FitNets: Hints for Thin Deep Nets
Romero, A., Ballas, N., Kahou, S. E., Chassang, A., Gatta, C., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1412.6550 · Cited by 2,859
Do Deep Nets Really Need to be Deep?
Ba, J. L., & Caruana, R. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1312.6184 · Cited by 1,335
Diving deeper into mentee networks
Venkatesan, R., & Li, B. (2016) · arXiv · doi:10.48550/arxiv.1604.08220 · Cited by 14 · cited 2×
“We study these KT techniques on three different types of studentteacher architectures: 1) Type I, where the student network is shallower than teacher network [2, 3, 6].”
BackgroundMethod
Research works printed but not linked: 1
- Geoffrey Hinton , Oriol Vinyals , and Jeff Dean . 2015. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 ( 2015 ). Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. 2015. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 (2015). · arXiv
Cited by
Convergence of Edge Computing and Deep Learning: A Comprehensive Survey
Wang, X., Han, Y., Leung, V. C. M., Niyato, D., Yan, X., & Chen, X. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2970550 · Cited by 1,498
Machine Learning at the Network Edge: A Survey
Murshed, M. G. S., Murphy, C., Hou, D., Khan, N., Ananthanarayanan, G., & Hussain, F. (2022) · ACM Computing Surveys · doi:10.1145/3469029 · Cited by 402
A survey on deep transfer learning to edge computing for mitigating the COVID-19 pandemic
Sufian, A., Ghosh, A., Sadiq, A. S., & Smarandache, F. (2020) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2020.101830 · Cited by 156
Jang, I., Kim, H., Lee, D., Son, Y.-S., & Kim, S. (2020) · IEEE Access · doi:10.1109/access.2020.3014922 · Cited by 51
Equilibrium in the Computing Continuum through Active Inference
Sedlak, B., Pujol, V. C., Donta, P. K., & Dustdar, S. (2024) · Future Generation Computer Systems · doi:10.1016/j.future.2024.05.056 · Cited by 47
E-Tree Learning: A Novel Decentralized Model Learning Framework for Edge AI
Yang, L., Lu, Y., Cao, J., Huang, J., & Zhang, M. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3052195 · Cited by 42
Artificial intelligence in edge devices
Singh, A., Saini, K., Nagar, V., Aseri, V., Sankhla, M. S., Pandit, P. P., & Chopade, R. L. (2022) · Advances in computers · doi:10.1016/bs.adcom.2022.02.013 · Cited by 22
Majewski, P., Mrzygłód, M., Lampa, P., Burduk, R., & Reiner, J. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107358 · Cited by 20
Terziyan, V. Y., & Tiihonen, T. (2024) · Procedia Computer Science · doi:10.1016/j.procs.2024.01.089 · Cited by 11
SCEH: Smart Customized E-Health Framework for Countryside Using Edge AI and Body Sensor Networks
Xu, C., Dong, M., Ota, K., Li, J., Yang, W., & Wu, J. (2019) · Globecom · doi:10.1109/globecom38437.2019.9014057 · Cited by 10
An Experimental Implementation of an Edge-based AI Engine with Edge-Cloud Coordination
Yamakami, T. (2018) · International Symposium on Communications and Information Technologies · doi:10.1109/iscit.2018.8587931 · Cited by 7
5G Network Edge Intelligence for Smart Operation and Maintenance of Offshore Wind Power
Gao, Y., Yang, L., Zhu, X., Jiang, C., Wang, H., You, S., & Xu, F. (2026) · Sensors · doi:10.3390/s26041390 · Cited by 4
Ding, B., Pan, Q., Li, Z., & Xu, X. (2025) · International Conference on Data Science and Intelligent Systems (DSIS) · doi:10.1109/dsis67228.2025.11390598 · Cited by 1
Machine Learning as Tag Estimation Method for ALOHA-based RFID system
Rodić, L. D., Stansic, I., Zovko, K., & Šolić, P. (2021) · International Conference on Smart and Sustainable Technologies (SpliTech) · doi:10.23919/splitech52315.2021.9566455 · Cited by 1
Smart and efficient sensor networks operation for 5G and beyond ecosystems
Sayed, A. E. (2024) · UPCommons institutional repository (Universitat Politècnica de Catalunya) · doi:10.5821/dissertation-2117-405945 · Cited by 1
Extending Knowledge Transfer in Data Analytics Through Cross-Domain AnalysisPreprint
Ghosh, M., Amrit, C., & Gromicho, J. A. S. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5016889 · Cited by 1
Traffic control for energy harvesting virtual small cells via reinforcement learning
Temesgene, D. A. (2020) · UPCommons institutional repository (Universitat Politècnica de Catalunya) · doi:10.5821/dissertation-2117-328205
Lessons Learned and Open Challenges
Wang, X., Han, Y., Leung, V. C. M., Niyato, D. T., Yan, X., & Chen, X. (2020) · Springer eBooks · doi:10.1007/978-981-15-6186-3_9
Chen, J.-M., Zhou, G.-D., Xi, J.-H., Yan, Y., & Liu, X.-W. (2026) · Mechanical Systems and Signal Processing · doi:10.1016/j.ymssp.2026.114692
Metadata
- Authors
- Sharma, R., Biookaghazadeh, S., & Zhao, M.
- Affiliations
- Arizona State University (US)
- Venue
- Proceedings of the 27th International Symposium on High-Performance Parallel and Distributed Computing
- Publisher
- ACM
- Published
- 2018-06-11
- Pages
- 15-16
- OpenAlex
- W2806329590
- Cited by
- 15
- References
- 5
- Citations by year
- 2026: 2 · 2025: 2 · 2024: 1 · 2023: 1 · 2022: 1 · 2021: 2 · 2020: 4 · 2019: 1 · 2018: 1
- FWCI
- 0.60
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Keywords
- knowledge transfer, knowledge distillation, edge computing, edge devices, deep learning, model compression, teacher-student models, convergence rate, accuracy, student architecture, transfer techniques
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.