Roadmap for edge AI: a Dagstuhl perspective
Open accessAbstract
Based on the collective input of Dagstuhl Seminar (21342), this paper presents a comprehensive discussion on AI methods and capabilities in the context of edge computing, referred as Edge AI. In a nutshell, we envision Edge AI to provide adaptation for data-driven applications, enhance network and radio access, and allow the creation, optimisation, and deployment of distributed AI/ML pipelines with given quality of experience, trust, security and privacy targets. The Edge AI community investigates novel ML methods for the edge computing environment, spanning multiple sub-fields of computer science, engineering and ICT. The goal is to share an envisioned roadmap that can bring together key actors and enablers to further advance the domain of Edge AI.
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?
Most survey articles on Edge AI were published before 2021 except [23], but [23] is only an overview or visionary work offering future directions in many relevant areas, not an in-depth study.
“Most of the survey articles on Edge AI have been published before 2021 except [23], but this [23] work is just an overview on Edge AI or a visionary work which offers only future directions in many relevant areas but not in-depth.”
From Edge AI: A survey (2023) · cites this paper as [23]
The Introduction and Section 5 describe the paper as an envisioned Edge AI roadmap spanning 5G, future cloud, and AI/ML, supporting its characterization as a broad, forward-looking work. But the claim that it offers only future directions and no in-depth discussion goes beyond the source: Section 3 says it will examine technological challenges in detail. The paper also does not establish when most other Edge AI surveys were published. (LLM, conf=0.85)
Experimental — accuracy not yet measuredEdge applications are leading the future generation of AI computing, which will positively affect everyday life at home, work, classroom, and travel.
“In reality, edge applications are leading the charge for the future generation of AI computing, which will have positive effects on our everyday lives in many contexts, including at home, in the workplace, in the classroom, and when travelling [23].”
From Edge AI: A survey (2023) · cites this paper as [23]
The Introduction describes Edge AI as a fast-evolving field already entering mainstream services, including connected vehicles and smart factories. Section 2.4 anticipates benefits for everyday life but says they still need evaluation. The paper does not establish that edge applications are leading the next generation of AI computing or support the full list of settings, particularly homes and classrooms. (LLM, conf=0.84)
Experimental — accuracy not yet measured
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
Data Poisoning Attacks Against Federated Learning Systems
Tolpegin, V., Truex, S., Gursoy, M. E., & Liu, L. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58951-6_24 · Cited by 927 · see citing works
Flagged first, then by citations
Research works printed but not linked: 9
- Setareh Maghsudi and Max Davy. 2021. Computational Models of Human Decision-Making with Application to the Internet of Everything. IEEE Wireless Communications 28, 1 (2021), 152-159.
- Setareh Maghsudi and Mihaela van der Schaar. 2019. Distributed Task Manage- ment in Cyber-Physical Systems: How to Cooperate Under Uncertainty? IEEE Trans. on Cognitive Communications and Networking 5, 1 (2019), 165-180.
- Nitinder Mohan, Lorenzo Corneo, Aleksandr Zavodovski, Suzan Bayhan, Walter Wong, and Jussi Kangasharju. 2020. Pruning Edge Research with Latency Shears. In Proceedings of the 19th ACM Workshop on Hot Topics in Networks (HotNets '20). 182-189.
- Md. Lushanur Rahman, J. Andrew Zhang, Kai Wu, Xiaojing Huang, Y. Jay Guo, Shanzhi Chen, and Jinhong Yuan. 2020. Enabling Joint Communication and Radio Sensing in Mobile Networks -A Survey. ArXiv abs/2006.07559 (2020). · arXiv
- Wiebke Toussaint and Aaron Yi Ding. 2020. Machine Learning Systems in the IoT: Trustworthiness Trade-offs for Edge Intelligence. In 2020 IEEE Second International Conference on Cognitive Machine Intelligence (CogMI). 177-184.
- Wiebke Toussaint, Akhil Mathur, Aaron Yi Ding, and Fahim Kawsar. 2021. Char- acterising the Role of Pre-Processing Parameters in Audio-based Embedded Machine Learning. In Proceedings of the 19th ACM Conference on Embedded Net- worked Sensor Systems (SenSys '21). ACM, 439-445.
- Blesson Varghese, Eyal de Lara, Aaron Yi Ding, Cheol-Ho Hong, Flavio Bonomi, Schahram Dustdar, Paul Harvey, Peter Hewkin, Weisong Shi, Mark Thiele, and Peter Willis. 2021. Revisiting the Arguments for Edge Computing Research. IEEE Internet Computing 25, 5 (2021), 36-42. https://doi.org/10.1109/MIC.2021.3093924 · DOI
- Joost Verbraeken, Matthijs Wolting, Jonathan Katzy, Jeroen Kloppenburg, Tim Verbelen, and Jan S. Rellermeyer. 2020. A Survey on Distributed Machine Learning. ACM Comput. Surv. 53, 2, Article 30 (March 2020), 33 pages.
- Q. Ye and Y. Zhang. 2016. Participation Behavior and Social Welfare in Repeated Task Allocations. In 2016 IEEE International Conference on Agents (ICA). 94-97.
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
- Authors
- Ding, A. Y., Peltonen, E., Meuser, T., Aral, A., Becker, C., Dustdar, S., Hiessl, T., Kranzlmüller, D., Liyanage, M., Maghsudi, S., Mohan, N., Ott, J., Rellermeyer, J. S., Schulte, S., Schulzrinne, H., Solmaz, G., Tarkoma, S., Varghese, B., & Wolf, L.
- Affiliations
- Delft University of Technology (NL)
- University of Oulu (FI)
- Technische Universität Darmstadt (DE)
- University of Vienna (AT)
- University of Mannheim (DE)
- TU Wien (AT)
- Ludwig-Maximilians-Universität München (DE)
- University Collage Dublin
- University of Tübingen (DE)
- Technical University of Munich (DE)
- Leibniz University Hannover (DE)
- Hamburg University of Technology (DE)
- Columbia University (US)
- NEC Labs Europe
- University of Helsinki (FI)
- University of St Andrews (GB)
- Technische Universität Braunschweig (DE)
- ORCID
- Venue
- ACM SIGCOMM Computer Communication Review
- Published
- 2022-01-30
- Volume
- 52
- Issue
- 1
- Pages
- 28-33
- ISSN
- 0146-4833 (ISSN-L)
- OpenAlex
- W4214509590
- Cited by
- 81
- References
- 1
- Citations by year
- 2026: 7 · 2025: 13 · 2024: 21 · 2023: 27 · 2022: 13
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- ACM SIGCOMM Computer Communication Review (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Zenodo (CERN European Organization for Nuclear Research) (submittedVersion, cc-by)
- Työväentutkimus Vuosikirja (submittedVersion)
- Research Repository UCD (University College Dublin) (submittedVersion, cc-by-nc-nd)
- tub.dok (Hamburg University of Technology) (submittedVersion)
- Research Repository (Delft University of Technology) (submittedVersion)
- MADOC (University of Mannheim) (submittedVersion)
- Data Archiving and Networked Services (DANS) (submittedVersion, other-oa)
- arXiv (Cornell University)
- Funders
- European Commission (HORIZON2020, 101021808)
- Austrian Science Fund (I 5201-N, I 5201)
- CHIST-ERA (CHIST-ERA-19-CES-005)
- Topics
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Computer Science › Information Systems › Internet of Things and AI
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- edge AI, edge computing, quality of experience, trust and security, privacy, data-driven applications, research roadmap
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
