Pith. sign in

REVIEW 4 major objections 6 minor 111 references

The paper argues that edge intelligence should be treated as a first-class, shareable entity, clustered semantically rather than by device, so that derived insights can be reused and combined across heterogeneous edge systems.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-01 08:56 UTC pith:6KHDJHKA

load-bearing objection A useful vision paper with a clean taxonomy and architecture; the reliability-gain motivation is a genuine untested assumption that should be reframed as an open hypothesis. the 4 major comments →

arxiv 2607.20937 v1 pith:6KHDJHKA submitted 2026-07-23 cs.AI cs.DC

Clustered Edge Intelligence: Beyond Just Convergence of Edge Computing and AI

Chinmaya Kumar Dehury , Boris Sedlak , Alaa Saleh , Ilir Murturi , Lauri Loven , Satish Narayana Srirama , Praveen Kumar Donta This is my paper
classification cs.AI cs.DC
keywords Clustered Edge Intelligenceedge computingintelligence-centric clusteringedge agentsedge-cloud continuumintelligence marketplaceknowledge graphscollaborative decision-making
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces Clustered Edge Intelligence (CEI), a vision for edge computing in which derived intelligence—not devices—is the unit of management. It distinguishes three forms of edge intelligence and focuses on the third: intelligence available at the edge, produced by edge agents from local data. CEI claims that semantically clustering such intelligences enables collaborative decision-making and cross-validation without sharing raw data, and proposes a three-layer architecture with intelligence inventories, discovery, observability, and marketplaces. If right, edge systems become more flexible, resilient, and less tied to specific hardware vendors.

Core claim

The central claim is that 'intelligence' derived at edge devices can be decoupled from the underlying hardware and treated as an independently manageable entity that can be uniquely identified, discovered, observed, shared, and reused. The paper contrasts device-centric clustering, where clusters of physical devices are managed and intelligence is tightly coupled to hardware, with intelligence-centric clustering, where semantically related intelligences are grouped at the edge controller level. It argues that clustering intelligences from multiple devices—for example, cross-checking CO2, temperature, humidity, and air-quality changes for fire detection—yields more reliable conclusions than a

What carries the argument

The key mechanism is intelligence-centric clustering (ICC), supported by edge agents—software entities that derive, publish, discover, and consume intelligence—and a three-layer architecture (edge device layer, edge controller layer, cloud layer) with an intelligence inventory and a discovery mechanism. The inventory maps each type of intelligence to the devices that can provide it, allowing queries by intelligence type rather than device identity, and it can list intelligences even when no connected device currently produces them.

Load-bearing premise

The framework is a vision without an implementation or evaluation, and its load-bearing assumption is that combining multiple independently derived intelligences yields more reliable conclusions than a single device's inference, and that semantically clustered intelligences will generally agree or be combinable.

What would settle it

A testbed experiment in which a fire-detection cluster cross-references CO2, temperature, humidity, and air-quality intelligences; if clustering does not reduce false positives or improve detection accuracy compared to a single CCTV-based detector across varied conditions, the central motivation for CEI weakens.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Edge systems could answer intelligence queries without knowing which device produced the answer, simplifying user interaction.
  • The same intelligence could be sourced from different devices at different times, reducing single-device dependency and vendor lock-in.
  • Collaborative decisions, such as fire detection or traffic congestion estimation, could be made by combining semantically related intelligence from heterogeneous devices without exchanging raw data.
  • An intelligence marketplace could emerge where intelligence is bought, sold, deployed, and reused across the edge-cloud continuum.
  • Lifecycle management of intelligence—publish, update, monitor, retire—would become a first-class operation alongside device management.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If intelligence becomes a first-class entity, trust and security attach to the intelligence itself rather than the device; confidence scores and provenance would need to be verifiable and standardized before cross-vendor sharing is practical.
  • A concrete, testable extension would compare fire-detection accuracy and false-alarm rates between single-source detection and multi-source intelligence clustering on a real testbed; the paper's motivation stands or falls on such a comparison.
  • CEI overlaps with federated learning in spirit but differs in substance: FL shares model updates, while CEI shares derived insights. Whether these approaches complement or compete remains unexplored in the paper.
  • Standardizing intelligence representation—identity, provenance, confidence, freshness, and communication requirements—is a prerequisite for any marketplace or large-scale interoperability, and would likely need to build on existing knowledge-graph and ontology work.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes Clustered Edge Intelligence (CEI), a vision and reference architecture in which the 'intelligence available at the edge' (EI-3F) is treated as a first-class, independently manageable entity, distinct from the two conventional meanings of edge intelligence: AI for edge resource management (EI-1F) and lightweight AI models deployed at the edge (EI-2F). The authors introduce edge agents, a three-layer architecture (edge device layer, edge controller layer, cloud environment layer), intelligence inventories, discovery and observability mechanisms, and an intelligence marketplace. The central proposal is to cluster intelligences semantically rather than clustering devices, enabling cross-device collaboration and reuse without sharing raw data. The paper compares device-centric (DCC) and intelligence-centric clustering (ICC), surveys baseline technologies and research dimensions, provides several use cases (fire detection, Industry 4.0, smart city, smart agriculture), and explicitly concludes that CEI is 'a vision and reference architecture rather than as an implemented and empirically evaluated system.'

Significance. If the central claims are supported, the paper would contribute a useful manageability layer for edge systems: decoupling intelligence from hardware, making it discoverable, reusable, and shareable, and enabling semantic clustering across heterogeneous devices. The taxonomy of the three forms of edge intelligence is clear, and the paper is honest about its status as a vision paper. The architecture is coherent and the survey of enabling technologies is broad. The main value is as a conceptual framework and research agenda. However, the core reliability/fusion assumption that combining multiple intelligences yields more reliable conclusions is not demonstrated, and some claims in the comparison table (notably security and privacy) are made without evidence. These gaps are load-bearing because they underpin the motivation for ICC and several stated benefits.

major comments (4)
  1. [§3.1, §3.2, §7.1–§7.3] The central motivation for ICC rests on the assumption stated in §3.1 that 'multiple types of information can be used to derive more reliable intelligence,' and the fire-detection example in §3.2 claims that cross-checking CCTV-based detection with CO2, temperature, humidity, and air-quality intelligence reduces false alarms. The use cases in §7 repeatedly assert that clusters can raise 'high-confidence' alerts. However, no fusion model is defined: there is no combination rule for confidence scores, no conflict-resolution strategy, no treatment of correlated errors, and no empirical or simulation-based validation. If conflicting or correlated intelligences do not improve accuracy, the primary value proposition of ICC weakens. The authors should either provide a formal or simulated demonstration of the fusion-reliability assumption or explicitly present it as an open hypothesis and soften
  2. [Table 3, §4.3, §4.4] The Security & privacy row of Table 3 states that ICC 'helps in enhancing security and privacy,' but the paper provides no security analysis or measurement to support this. Section 4.3 lists a Regulator and a Sharing Mechanism at the cloud layer, and Section 4.4 mentions 'Inter-Agent Communication Protocols,' but no concrete security mechanism or privacy-preserving property is defined. Sharing derived intelligence can itself leak sensitive information or enable inference attacks, so the claimed benefit is not self-evident. This claim should either be substantiated with a concrete mechanism and analysis, or qualified as a potential (unverified) advantage.
  3. [§5.6 vs. §8.3, §3.3] Section 5.6 states that 'to the best of our knowledge, no specific clustering algorithm is developed to perform the same for intelligence,' yet Section 8.3 cites prior work that appears to do exactly this: reference [47] is credited with forming 'logical clusters based on the specific intelligence devices possess,' and reference [106] is said to allow nodes 'to cluster autonomously based on semantic relevance.' In addition, §3.3 begins by saying this section is 'in continuation' of the authors' own previous work [47]. The paper must clearly delineate what is new relative to CCEI-IoT and other cited works. As written, the novelty of the intelligence-centric clustering contribution is not fully established.
  4. [§9 and general evaluation] The paper candidly states that CEI is 'presented as a vision and reference architecture rather than as an implemented and empirically evaluated system.' This is a significant limitation for a journal article. The operational feasibility of the architecture—intelligence inventory management, semantic discovery, observability overhead, and cluster formation in resource-constrained edge environments—is entirely untested. Even a small proof-of-concept or simulation of one component (e.g., intelligence discovery or the proposed fire-detection clustering) would considerably strengthen the paper and turn the vision into a testable framework. In the current form, the central claims remain plausible but unverified.
minor comments (6)
  1. [Abstract/§1 contributions] Typo: 'intelliegnce' should be 'intelligence.' Also, the contribution list repeats the introductory paragraph almost verbatim; consider tightening the wording.
  2. [§4.2] The text says 'The answer lies in Figure 5,' but Figure 5 shows the DCC/ICC comparison, not the overall architecture. The cross-reference should be to Figure 6.
  3. [Table 3] The Management row contains typos: 'Intelligence wil not be affected' and 'Devices wil not be affected.' Also, the Security & privacy row needs a reference to a section that actually supports the claim (see major comment).
  4. [§5.4.1 and §5.4.3] The tool name is spelled 'Hysterix' but the standard spelling is 'Hystrix.' In §5.4.3, 'Unlink DD' should be 'Unlike DD.'
  5. [§5.1, §5.6, §6.3] Typos: 'linked ilst' should be 'linked list'; 'inteligence' should be 'intelligence'; 'aumation' should be 'automation.'
  6. [§8.3] The concluding sentence says 'DCC emerges as a powerful enabler for CEI,' but in context this appears to refer to intelligence-centric clustering (ICC) or the cited works [106,107], not device-centric clustering DCC, which is the approach being contrasted. Please correct.

Circularity Check

0 steps flagged

No circular derivation: the paper is an explicitly labeled vision/reference architecture, with no fitted-input-as-prediction, no uniqueness theorem, and no definitional reduction. The self-citation to prior CCEI-IoT work is acknowledged background, not a load-bearing proof step.

full rationale

The paper does not present a mathematical derivation chain, a fitted model, or an empirical prediction. Its central claim—that derived intelligence should be treated as a first-class, shareable, clusterable entity—is a conceptual/architectural proposal, explicitly framed as 'a vision and reference architecture rather than as an implemented and empirically evaluated system' (Section 9). The fire-detection and traffic examples are illustrative scenarios, not computed predictions, and the statement in Section 3.1 that 'multiple types of information can be used to derive more reliable intelligence' is presented as a belief/assumption, not as a result derived from the CEI definitions. The only self-citation in the load-bearing vicinity is Section 3.3's phrase 'our previous work on clustered and cohesive edge intelligence [47]' and Section 8.3's attribution to Dehury et al. [47] for forming logical clusters. This is an honest continuation/attribution statement rather than a circular justification: the paper does not argue that CEI is valid because [47] says so, and it does not claim a unique or externally forced choice backed by a self-authored theorem. The absence of an empirical fusion-reliability test is a support gap and a correctness risk, but not circularity, because the reliability claim is stated as an untested assumption, not as a derived output of the framework. No quoted equation or fitted parameter reduces to an input, so no circular step meets the evidentiary bar required by the review rules.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 3 invented entities

No numeric parameters are fitted anywhere in the paper. The central claim rests on domain assumptions about the representability and composability of intelligence, plus the use of existing clustering/ontology tools that have not been validated for this setting. The invented entities are architectural components, not physical entities; none has a falsifiable handle outside the paper.

axioms (3)
  • domain assumption Derived intelligence can be represented as a standalone semantic object independent of the device/data that produced it (with metadata, access URL, confidence, etc.).
    Section 3 states the edge agent 'encapsulates these components and decouples them from the host device'; Section 5.1.1 introduces an 'Intelligence Layer'. If intelligence cannot be abstracted from hardware, the ICC model collapses.
  • domain assumption Semantically clustering multiple intelligences yields more reliable/accurate conclusions than single-source inference.
    Section 3.1: 'we believe that multiple types of information can be used to derive more reliable intelligence'; all use cases in Section 7 depend on this premise.
  • domain assumption Existing clustering algorithms, knowledge graphs, and formats such as OKF can be adapted to intelligence artifacts.
    Section 5.6 says existing clustering mechanisms 'can be used as the baseline'; Section 5.2 suggests OKF 'may complement' CEI. No proof or prototype is provided.
invented entities (3)
  • Edge Agent (EA) no independent evidence
    purpose: Autonomous software on an edge device that derives, publishes, discovers, consumes, and clusters intelligence.
    Defined conceptually in Section 3; no prototype, formal spec, or falsifiable prediction.
  • Intelligence Inventory no independent evidence
    purpose: Registry at edge controller or cloud mapping intelligences to devices and metadata, enabling device-independent discovery.
    Proposed architecture component (Section 4.2); not implemented.
  • Edge Intelligence Marketplace (EIM) no independent evidence
    purpose: Platform where intelligence producers and consumers buy, sell, or exchange AI models, rules, or derived intelligence.
    Conceptual extension of model hubs/e-commerce to edge intelligence (Section 6.1); no system or business model implemented.

pith-pipeline@v1.3.0-alltime-deepseek · 33371 in / 15043 out tokens · 148073 ms · 2026-08-01T08:56:45.490910+00:00 · methodology

0 comments
read the original abstract

We are moving from an information age to the age of intelligence. A decade, or possibly less than that, data will not be the gold anymore rather the derived intelligence out of the data and the information we posses from the edge of the network. Existing Edge Intelligence research focuses mainly on two directions: using AI for edge resource management and deploying lightweight AI models on edge devices. However, existing edge computing research lacks an intelligence-centric framework in which derived intelligence is treated as a first-class, independently manageable entity that can be described, discovered, observed, shared, reused, and dynamically clustered across heterogeneous edge devices and applications. To address these research gaps, we introduced Clustered Edge Intelligence, a visionary intelligence-centric approach. The aim of CEI is to make intelligence a shareable and reusable first-class entity that can be independently represented, discovered, observed, exchanged, and managed across the distributed edge-cloud continuum. We present a three layer CEI architecture and examine enabling technologies and research dimensions, including intelligence inventories, semantic knowledge representation, communication, discoverability, observability, lifecycle automation, clustering mechanisms, marketplaces, interoperability, and standardization.

Figures

Figures reproduced from arXiv: 2607.20937 by Alaa Saleh, Boris Sedlak, Chinmaya Kumar Dehury, Ilir Murturi, Lauri Loven, Praveen Kumar Donta, Satish Narayana Srirama.

Figure 1
Figure 1. Figure 1: In this example, data is first collected from [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 1
Figure 1. Figure 1: The transformation from Data to Intelligence: an example in Edge device [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Organization of the manuscript For example, edge devices such as roadside CCTV an￾alytics units or Jetson-based video processors often un￾dergo thermal stress during continuous high-rate infer￾ence [23]. A TinyML-based regression model can be em￾bedded locally to predict short-term thermal load (e.g., 30–60 seconds ahead) based on current CPU/GPU utiliza￾tion, ambient temperature, and inference rate. When … view at source ↗
Figure 3
Figure 3. Figure 3: Different forms of Edge Intelligence. 2.3. EI-3F: Intelligence available at edge However, unlike the above-mentioned meanings of “edge intelligence”, it can also be referred to as the “intelligence available at the edge of the network”, also known as 3rd form of EI. Some of the examples include intelligence re￾lated to energy production by Photovoltaic panels deployed atop city buses, intelligence related … view at source ↗
Figure 4
Figure 4. Figure 4: Comparison between traditional and CEI-enabled fire-detection mechanisms (an example): (a) single-source detection, (b) [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: A pictorial representation of (a) Device-centric clustering approach, and (b) intelligence-centric clustering approach. [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: illustrates the three-tiered architecture of Clus￾tered Edge Intelligence (CEI), consisting of edge devices at the foundational layer, an edge gateway or controller in the intermediary layer, and the cloud at the top. The architecture’s core objective is the effective management of intelligence within the edge infrastructure. Edge de￾vices exhibit two forms of intelligence: Core intelligence, derived from … view at source ↗
Figure 7
Figure 7. Figure 7: A detailed view of edge device in CEI. or even millions of edge controllers. The cloud environ￾ment at the top, is equipped with a global view and is responsible for managing and coordinating all edge con￾trollers. Similar to an edge controller, the cloud layer also consists of an Intelligence Inventory and a Discov￾ery Mechanism components. The discovery mechanism enables an edge controller to identify an… view at source ↗
Figure 8
Figure 8. Figure 8: Different layers of abstraction in an edge device. [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: A general architecture of service discovery [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Interaction of stakeholders within Intelligence Market [PITH_FULL_IMAGE:figures/full_fig_p018_10.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

111 extracted references · 13 canonical work pages · 1 internal anchor

  1. [1]

    Rowley, The wisdom hierarchy: representations of the dikw hierarchy, Journal of information science 33 (2) (2007) 163–180

    J. Rowley, The wisdom hierarchy: representations of the dikw hierarchy, Journal of information science 33 (2) (2007) 163–180

    2007

  2. [2]

    S. J. Russell, Artificial intelligence a modern ap- proach, Pearson Education, Inc., 2010

    2010

  3. [3]

    Ieee standard glossary of software engineering terminology, IEEE Std 610.12-1990 (1990) 1– 84doi:10.1109/IEEESTD.1990.101064

    arXiv 1990

  4. [4]

    Sheth, C

    A. Sheth, C. Henson, S. S. Sahoo, Semantic sensor web, IEEE Internet computing 12 (4) (2008) 78–83. 22

    2008

  5. [5]

    C. K. Gately, L. R. Hutyra, S. Peterson, I. S. Wing, Urban emissions hotspots: Quantifying vehicle con- gestion and air pollution using mobile phone gps data, Environmental pollution 229 (2017) 496–504

    2017

  6. [6]

    Espadaler-Clapés, E

    J. Espadaler-Clapés, E. Barmpounakis, N. Geroli- minis, Traffic congestion and noise emissions with detailed vehicle trajectories from uavs, Transporta- tion Research Part D: Transport and Environment 121 (2023) 103822

    2023

  7. [7]

    J. Du, L. Zhao, J. Feng, X. Chu, Compu- tation offloading and resource allocation in mixed Fog/Cloud computing systems with min-max fairness guarantee, IEEE Transac- tions on Communications 66 (4) (2018) 1594–1608. doi:10.1109/TCOMM.2017.2787700

    arXiv 2018

  8. [8]

    Achkouty, L

    F. Achkouty, L. Gallon, R. Chbeir, Rdsc: Range- based device spatial clustering for iot networks, Sen- sors 24 (17) (2024) 5851

    2024

  9. [9]

    A. K. Takele, B. Villányi, Resource-efficient clus- tered federated learning framework for industry 4.0 edge devices, Ai 6 (2) (2025) 30

    2025

  10. [10]

    B. A. Begum, S. V. Nandury, Data aggregation pro- tocols for wsn and iot applications–a comprehensive survey, Journal of King Saud University-Computer and Information Sciences 35 (2) (2023) 651–681

    2023

  11. [11]

    Ghrab, I

    S. Ghrab, I. Lahyani, S. Yangui, M. Jmaiel, A core iot ontology for automation support in edge comput- ing, Service Oriented Computing and Applications 17 (1) (2023) 25–37

    2023

  12. [12]

    Singh, S

    R. Singh, S. S. Gill, Edge ai: a survey, Internet of Things and Cyber-Physical Systems 3 (2023) 71–92

    2023

  13. [13]

    Choudhury, M

    A. Choudhury, M. Ghose, A. Islam, et al., Machine learning-based computation offloading in multi- accessedgecomputing: Asurvey, JournalofSystems Architecture 148 (2024) 103090

    2024

  14. [14]

    Sandler, A

    M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, L.- C. Chen, Mobilenetv2: Inverted residuals and linear bottlenecks, in: Proceedings of the IEEE conference on computer vision and pattern recognition, 2018, pp. 4510–4520

    2018

  15. [15]

    F. N. Iandola, S. Han, M. W. Moskewicz, K. Ashraf, W. J. Dally, K. Keutzer, SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size, arXiv preprint arXiv:1602.07360 (2016)

    Pith/arXiv arXiv 2016

  16. [16]

    Zhang, X

    X. Zhang, X. Zhou, M. Lin, J. Sun, Shufflenet: An extremely efficient convolutional neural network for mobile devices, in: Proceedings of the IEEE con- ference on computer vision and pattern recognition, 2018, pp. 6848–6856

    2018

  17. [17]

    D. Vyas, M. d. Prado, T. Verbelen, Towards smart and adaptive agents for active sensing on edge de- vices (Jan. 2025). doi:10.48550/arXiv.2501.06262

    work page doi:10.48550/arxiv.2501.06262 2025

  18. [18]

    K. S. Velaga, Y. Guo, W. Yu, Edge ai for smart cities: Foundations, challenges, and opportunities, Smart Cities 8 (6) (2025) 211

    2025

  19. [19]

    Y. Hu, Q. Li, Y. Chai, D. Wu, L. Lu, N. Shi, Y. Teng, Y. Zhang, Ai service deploy- ment and resource allocation optimization based on human-like networking architecture, IEEE Inter- net of Things Journal 11 (14) (2024) 24795–24813. doi:10.1109/JIOT.2024.3384546

    arXiv 2024

  20. [20]

    Y. Xie, Q. Fang, An energy-aware gener- ative ai edge inference framework for low- power iot devices, Electronics 14 (20) (2025). doi:10.3390/electronics14204086. URLhttps://www.mdpi.com/2079-9292/14/20/ 4086

    work page doi:10.3390/electronics14204086 2025

  21. [21]

    Shah, M.A

    S.D.A. Shah, M.A. Gregory, F. Bouhafs, F. D. Har- tog, Artificial intelligence-defined wireless network- ing for computational offloading and resource alloca- tion in edge computing networks, IEEE Open Jour- nal of the Communications Society 5 (2024) 2039–

    2024

  22. [22]

    C. K. Dehury, S. Poojara, S. N. Srirama, Def- drel: Towards a sustainable serverless func- tions deployment strategy for fog-cloud envi- ronments using deep reinforcement learning, Applied Soft Computing 152 (2024) 111179. doi:https://doi.org/10.1016/j.asoc.2023.111179. URLhttps://www.sciencedirect.com/science/ article/pii/S1568494623011973

    arXiv 2024

  23. [23]

    Nisha, R

    P. Nisha, R. Vinay, K. Laad, P. Sasmal, T. Haraki, C. Juyal, M. A. Gabir Elbakri, A. Acharyya, Run-time prevention of thermal throttling on the edge using reinforcement-learning based predictive thermal aware power and performance manage- ment, in: 2024 22nd IEEE Interregional NEW- CAS Conference (NEWCAS), 2024, pp. 273–277. doi:10.1109/NewCAS58973.2024.10666109

    arXiv 2024

  24. [24]

    Sedlak, P

    B. Sedlak, P. Raith, A. Morichetta, V. C. Pu- jol, S. Dustdar, Multi-Dimensional Autoscaling of Stream Processing Services on Edge Devices (Oct. 2025). doi:10.48550/arXiv.2510.06882

    work page doi:10.48550/arxiv.2510.06882 2025

  25. [25]

    Pandith, M

    O. Pandith, M. K. Pandit, A. Gujarati, A. Mehta, R. Sen, Iceedge: Thermal-aware machine learn- ing inference serving for emerging edge appli- cations, ACM Trans. Sen. Netw. (Mar. 2026). doi:10.1145/3801093. URLhttps://doi.org/10.1145/3801093 23

    work page doi:10.1145/3801093 2026

  26. [26]

    Mazloomi, H

    A. Mazloomi, H. Sami, J. Bentahar, H. Otrok, A. Mourad, Reinforcement learning framework for server placement and workload allocation in multiaccess edge computing, IEEE Internet of Things Journal 10 (2) (2023) 1376–1390. doi:10.1109/JIOT.2022.3205051

    arXiv 2023

  27. [27]

    C. K. Dehury, S. N. Srirama, An efficient service dispersal mechanism for fog and cloud computing using deep reinforcement learning, in: 2020 20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID), 2020, pp. 589–598. doi:10.1109/CCGrid49817.2020.00-34

    arXiv 2020

  28. [28]

    Chang, R

    Y. Chang, R. Thompson, C. Hixenbaugh, Tiny long- short term memory model for resource-constrained prediction of battery cycle life, in: G. Nicosia, V. Ojha, S. Giesselbach, M. P. Pardalos, R. Umeton (Eds.), Machine Learning, Optimization, and Data Science, Springer Nature Switzerland, Cham, 2025, pp. 131–144

    2025

  29. [29]

    Albanese, M

    A. Albanese, M. Nardello, D. Brunelli, Automated Pest Detection With DNN on the Edge for Preci- sion Agriculture, IEEE Journal on Emerging and Se- lected Topics in Circuits and Systems 11 (3) (2021) 458–467, conferenceName: IEEEJournalonEmerg- ing and Selected Topics in Circuits and Systems. doi:10.1109/JETCAS.2021.3101740

    arXiv 2021

  30. [30]

    D. Aghi, S. Cerrato, V. Mazzia, M. Chiaberge, Deep Semantic Segmentation at the Edge for Autonomous Navigation in Vineyard Rows, in: 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2021, pp. 3421–3428, iSSN: 2153-

    2021

  31. [31]

    Radanliev, D

    P. Radanliev, D. De Roure, K. Page, J. R. C. Nurse, R. Mantilla Montalvo, O. Santos, L. Maddox, P. Burnap, Cyber risk at the edge: current and future trends on cyber risk analytics and artificial intelligence in the industrial internet of things and industry 4.0 supply chains, Cybersecurity 3 (1) (2020) 13. doi:10.1186/s42400-020-00052-8. URLhttps://doi.o...

    work page doi:10.1186/s42400-020-00052-8 2020

  32. [32]

    S. N. Srirama, Fog computing and large language models: A vision for mutual beneficiaries, IEEE Computer (2026). doi:10.1109/MC.2026.3708686

    arXiv 2026

  33. [33]

    Sedlak, A

    B. Sedlak, A. Morichetta, Y. Wang, Y. Fei, L. Wang, S. Dustdar, X. Qu, SLO-Aware Task Of- floading Within Collaborative Vehicle Platoons, in: W. Gaaloul, M. Sheng, Q. Yu, S. Yangui (Eds.), Service-Oriented Computing, Springer Nature, Sin- gapore, 2024, pp. 72–86

    2024

  34. [34]

    C. Qiu, X. Wang, H. Yao, Z. Xiong, F. R. Yu, V. C. M. Leung, Bring Intelligence among Edges: A Blockchain-Assisted Edge Intelligence Approach, in: GLOBECOM 2020 - 2020 IEEE Global Communi- cations Conference, 2020, pp. 1–6, iSSN: 2576-6813. doi:10.1109/GLOBECOM42002.2020.9348271

    arXiv 2020

  35. [35]

    Rausch, S

    T. Rausch, S. Dustdar, Edge Intelligence: The Convergence of Humans, Things, and AI, in: 2019 IEEE International Conference on Cloud Engineering (IC2E), 2019, pp. 86–96. doi:10.1109/IC2E.2019.00022

    arXiv 2019

  36. [36]

    Plastiras, M

    G. Plastiras, M. Terzi, C. Kyrkou, T. Theocharidcs, Edge Intelligence: Challenges and Opportunities of Near-Sensor Machine Learning Applications, in: 2018 IEEE 29th International Conference on Application-specificSystems, ArchitecturesandPro- cessors (ASAP), 2018, pp. 1–7, iSSN: 2160-052X. doi:10.1109/ASAP.2018.8445118

    arXiv 2018

  37. [37]

    6G White Paper on Edge Intelligence

    E. Peltonen, M. Bennis, M. Capobianco, M. Deb- bah, A. Ding, F. Gil-Castiñeira, M. Jurmu, T. Kar- vonen, M. Kelanti, A. Kliks, T. Leppänen, L. Lovén, T. Mikkonen, A. Rao, S. Samarakoon, K. Seppänen, P. Sroka, S. Tarkoma, T. Yang, 6G White Paper on Edge Intelligence, arXiv:2004.14850 [cs] (Apr. 2020). doi:10.48550/arXiv.2004.14850. URLhttp://arxiv.org/abs/...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2004.14850 2004

  38. [38]

    D. N. Molokomme, A. J. Onumanyi, A. M. Abu- Mahfouz, Edge Intelligence in Smart Grids: A Sur- vey on Architectures, Offloading Models, Cyber Se- curity Measures, and Challenges, Journal of Sensor and Actuator Networks 11 (3) (2022) 47, number: 3 Publisher: Multidisciplinary Digital Publishing In- stitute. doi:10.3390/jsan11030047. URLhttps://www.mdpi.com/...

    work page doi:10.3390/jsan11030047 2022

  39. [39]

    Parekh, K

    B. Parekh, K. Amin, Edge Intelligence: A Robust Reinforcement of Edge Computing and Artificial In- telligence, in: P. K. Singh, Z. Polkowski, S. Tan- war, S. K. Pandey, G. Matei, D. Pirvu (Eds.), In- novations in Information and Communication Tech- nologies (IICT-2020), Advances in Science, Technol- ogy&Innovation, SpringerInternationalPublishing, Cham, 2...

    2020

  40. [40]

    Sedlak, I

    B. Sedlak, I. Murturi, P. K. Donta, S. Dust- dar, A Privacy Enforcing Framework for Trans- forming Data Streams on the Edge, IEEE Trans- actions on Emerging Topics in Computing (2023). doi:10.1109/TETC.2023.3315131

    arXiv 2023

  41. [41]

    Thakur, S

    P. Thakur, S. Goel, E. Puthooran, Edge ai enabled iot framework for secure smart home infrastructure, Procedia Computer Science 235 (2024) 3369–3378

    2024

  42. [42]

    Savaglio, P

    C. Savaglio, P. Mazzei, G. Fortino, Edge intelligence for industrial iot: Opportunities and limitations, Procedia Computer Science 232 (2024) 397–405. 24

    2024

  43. [43]

    T. Shen, L. Zhang, R. Geng, S. Li, B. Sun, Traf- fic prediction for diverse edge iot data using graph network, Journal of Cloud Computing 13 (1) (2024) 84

    2024

  44. [44]

    W. Qi, Q. Li, Q. Song, L. Guo, A. Jamalipour, Extensive edge intelligence for future vehicular net- works in 6g, IEEE Wireless Communications 28 (4) (2021) 128–135

    2021

  45. [45]

    McMahan, E

    B. McMahan, E. Moore, D. Ramage, S. Hampson, B. A. y Arcas, Communication-efficient learning of deep networks from decentralized data, in: Artifi- cial intelligence and statistics, Pmlr, 2017, pp. 1273– 1282

    2017

  46. [46]

    Y. Li, X. Wang, R. Zeng, P. K. Donta, I. Murturi, M. Huang, S. Dustdar, Federated domain general- ization: A survey, Proceedings of the IEEE (2025)

    2025

  47. [47]

    C. K. Dehury, P. K. Donta, S. Dustdar, S. N. Sri- rama, CCEI-IoT: Clustered and cohesive edge intel- ligence in Internet of Things, in: 2022 IEEE Inter- national Conference on Edge Computing and Com- munications (EDGE), IEEE, 2022, pp. 33–40

    2022

  48. [48]

    Silva, T

    M. Silva, T. Gomes, M. Ekpanyapong, A. Tavares, S. Pinto, Chameliot: A tightly-and loosely-coupled hardware-assisted os framework for low-end iot de- vices, Real-Time Systems 60 (1) (2024) 150–196

    2024

  49. [49]

    R. Chan, W. K. Yan, J. M. Ma, K. M. Loh, T. Yu, M.Y.H.Low, K.P.Yar, H.Rehman, T.C.Phua, Iot devices deployment challenges and studies in build- ing management system, Frontiers in the Internet of Things 2 (2023) 1254160

    2023

  50. [50]

    Höglund, S

    J. Höglund, S. Bouget, M. Furuhed, J. Preuß Matts- son, G. Selander, S. Raza, Autopki: public key in- frastructure for iot with automated trust transfer, International Journal of Information Security 23 (3) (2024) 1859–1875

    2024

  51. [51]

    Pathak, K

    M. Pathak, K. N. Mishra, S. P. Singh, Securing data and preserving privacy in cloud iot-based technolo- gies an analysis of assessing threats and develop- ing effective safeguard, Artificial intelligence review 57 (10) (2024) 269

    2024

  52. [52]

    Dib, Communication-efficient agentic edge intelligence: A survey of federated multi- agent information fusion for llm-driven edge ai, Information Fusion 136 (2026) 104577

    O. Dib, Communication-efficient agentic edge intelligence: A survey of federated multi- agent information fusion for llm-driven edge ai, Information Fusion 136 (2026) 104577. doi:https://doi.org/10.1016/j.inffus.2026.104577

    arXiv 2026

  53. [53]

    Wagner, F

    S. Wagner, F. Deissenboeck, Abstractness, speci- ficity, and complexity in software design, in: Pro- ceedings of the 2nd International Workshop on the Role of Abstraction in Software Engineering, 2008, pp. 35–42

    2008

  54. [54]

    Simmann, V

    G. Simmann, V. Veeranna, R. Kriesten, Design of an alternative hardware abstraction layer for embed- ded systems with time-controlled hardware access, in: 2024 Stuttgart International Symposium, SAE Technical Paper, 2024

    2024

  55. [55]

    Lercher, J

    A. Lercher, J. Glock, C. Macho, M. Pinzger, Mi- croservice api evolution in practice: A study on strategies and challenges, Journal of Systems and Software 215 (2024) 112110

    2024

  56. [56]

    Bartolini, M

    I. Bartolini, M. Patella, A stream processing ab- straction framework, Frontiers in Big Data 6 (2023) 1227156

    2023

  57. [57]

    Silva, J

    M. Silva, J. Santos, M. Curado, The path towards virtualized wireless communications: A survey and research challenges, Journal of Network and Systems Management 32 (1) (2024) 12

    2024

  58. [58]

    Krishnan, K

    P. Krishnan, K. Jain, A. Aldweesh, P. Prabu, R. Buyya, Openstackdp: a scalable network security framework for sdn-based openstack cloud infrastruc- ture, Journal of Cloud Computing 12 (1) (2023) 26

    2023

  59. [59]

    W. R. Cook, On understanding data abstraction, re- visited, in: Proceedings of the 24th ACM SIGPLAN conference on Object oriented programming systems languages and applications, 2009, pp. 557–572

    2009

  60. [60]

    Shirali, M

    M. Shirali, M. F. Sani, Z. Ahmadi, E. Serral, Llm- based event abstraction and integration for iot- sourced logs, in: International Conference on Busi- ness Process Management, Springer, 2024, pp. 138– 149

    2024

  61. [61]

    Villagrossi, M

    E. Villagrossi, M. Delledonne, M. Faroni, M. Beschi, N. Pedrocchi, Hiding task-oriented programming complexity: an industrial case study, Interna- tional Journal of Computer Integrated Manufactur- ing 36 (11) (2023) 1629–1648

    2023

  62. [62]

    L.Kong, J.Tan, J.Huang, G.Chen, S.Wang, X.Jin, P. Zeng, M. Khan, S. K. Das, Edge-computing- driven internet of things: A survey, ACM computing surveys 55 (8) (2022) 1–41

    2022

  63. [63]

    Hogan, E

    A. Hogan, E. Blomqvist, M. Cochez, C. d’Amato, G. D. Melo, C. Gutierrez, S. Kirrane, J. E. L. Gayo, R. Navigli, S. Neumaier, et al., Knowledge graphs, ACM Computing Surveys (Csur) 54 (4) (2021) 1–37

    2021

  64. [64]

    C. K. Dehury, S. N. Srirama, Mastering DevOps: A Cloud Engineering and Data Science Perspective, Morgan Kaufmann, 2026. doi:https://doi.org/ 10.1016/C2025-0-00512-4

    work page doi:10.1016/c2025-0-00512-4 2026

  65. [65]

    Dauda, O

    A. Dauda, O. Flauzac, F. Nolot, A survey on iot ap- plication architectures, Sensors 24 (16) (2024) 5320. 25

    2024

  66. [66]

    T. F. D. B. Sarasola, A. García, J. L. Ferrando, Iiot protocolsforedge/fog andcloudcomputinginindus- trial ai: A high frequency perspective, International Journal of Cloud Applications and Computing (IJ- CAC) 14 (1) (2024) 1–30

    2024

  67. [67]

    Lazidis, K

    A. Lazidis, K. Tsakos, E. G. Petrakis, Publish– subscribeapproachesfortheiotandthecloud: Func- tional and performance evaluation of open-source systems, Internet of Things 19 (2022) 100538

    2022

  68. [68]

    Gupta, D

    R. Gupta, D. Reebadiya, S. Tanwar, 6g-enabled edge intelligence for ultra-reliable low latency applica- tions: Vision and mission, Computer Standards & Interfaces 77 (2021) 103521

    2021

  69. [69]

    Huang, Q

    K. Huang, Q. Lan, Z. Liu, L. Yang, Semantic data sourcing for 6g edge intelligence, IEEE Communica- tions Magazine 61 (12) (2023) 70–76

    2023

  70. [70]

    Murturi, C

    I. Murturi, C. Avasalcai, C. Tsigkanos, S. Dust- dar, Edge-to-edge resource discovery using metadata replication, in: 2019 IEEE 3rd International Confer- ence on Fog and Edge Computing (ICFEC), IEEE, 2019, pp. 1–6

    2019

  71. [71]

    Murturi, S

    I. Murturi, S. Dustdar, A Decentralized Approach for Resource Discovery using Metadata Replication in Edge Networks, IEEE Transactions on Services Computing 15 (5) (2022) 2526–2537

    2022

  72. [72]

    Dragoni, S

    N. Dragoni, S. Giallorenzo, A. L. Lafuente, M. Maz- zara, F. Montesi, R. Mustafin, L. Safina, Microser- vices: yesterday, today, and tomorrow, Present and ulterior software engineering (2017) 195–216

    2017

  73. [73]

    Burns, B

    B. Burns, B. Grant, D. Oppenheimer, E. Brewer, J. Wilkes, Borg, omega, and kubernetes, Communi- cations of the ACM 59 (5) (2016) 50–57

    2016

  74. [74]

    S. E. Czerwinski, B. Y. Zhao, T. D. Hodes, A. D. Joseph, R. H. Katz, An architecture for a secure service discovery service, in: Proceed- ings of the 5th annual ACM/IEEE international conference on Mobile computing and networking, ACM, Seattle Washington USA, 1999, pp. 24–35. doi:10.1145/313451.313462. URLhttps://dl.acm.org/doi/10.1145/313451. 313462

    arXiv 1999

  75. [75]

    Hayat, R

    O. Hayat, R. Ngah, S. Z. Mohd Hashim, M. H. Dahri, R. Firsandaya Malik, Y. Rahayu, Device Dis- covery in D2D Communication: A Survey, IEEE Access 7 (2019) 131114–131134, conference Name: IEEE Access. doi:10.1109/ACCESS.2019.2941138

    arXiv 2019

  76. [76]

    Baccelli, N

    F. Baccelli, N. Khude, R. Laroia, J. Li, T. Richard- son, S. Shakkottai, S. Tavildar, X. Wu, On the design of device-to-device autonomous discov- ery, in: 2012 Fourth International Conference on Communication Systems and Networks (COM- SNETS 2012), 2012, pp. 1–9, iSSN: 2155-2509. doi:10.1109/COMSNETS.2012.6151335

    arXiv 2012

  77. [77]

    P. C. Ccori, L. C. C. De Biase, M. K. Zuffo, F. S. C. da Silva, Device discovery strategies for the IoT, in: 2016 IEEE International Symposium on Consumer Electronics (ISCE), 2016, pp. 97–98, iSSN: 2159-

    2016

  78. [78]

    Beerliova, F

    Z. Beerliova, F. Eberhard, T. Erlebach, A. Hall, M. Hoffmann, M. Mihal’ak, L. S. Ram, Network Dis- covery and Verification, IEEE Journal on Selected Areas in Communications 24 (12) (2006) 2168–2181, conference Name: IEEE Journal on Selected Areas inCommunications. doi:10.1109/JSAC.2006.884015

    arXiv 2006

  79. [79]

    R. I. Ansari, C. Chrysostomou, S. A. Hassan, M. Guizani, S. Mumtaz, J. Rodriguez, J. J. P. C. Ro- drigues, 5G D2D Networks: Techniques, Challenges, and Future Prospects, IEEE Systems Journal 12 (4) (2018) 3970–3984, conference Name: IEEE Systems Journal. doi:10.1109/JSYST.2017.2773633

    arXiv 2018

  80. [80]

    Kosińska, B

    J. Kosińska, B. Baliś, M. Konieczny, M. Malawski, S. Zieliński, Toward the observability of cloud-native applications: The overview of the state-of-the-art, IEEE Access 11 (2023) 73036–73052

    2023

Showing first 80 references.