Pith. sign in

REVIEW 4 major objections 5 minor 44 references

Lyapunov Stability-Aware Stackelberg Game for Low-Altitude Economy: A Control-Oriented Pruning-Based DRL Approach

T0 review · 4 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

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

Pith's one-line read UAV control stability can be converted into a bandwidth price floor, and a pruning-based reinforcement learning algorithm can reach the resulting market equilibrium on resource-limited hardware.

desk verdict Useful systems integration with a real proof gap: the Lyapunov-to-bandwidth mapping in Proposition 1 conflates per-user expected latency with the expected max, and the state-dependent threshold is not handled, so the stability guarantee is not established as stated. read the letter →

arxiv 2602.01131 v2 pith:OYV6OBRL submitted 2026-02-01 cs.AI

Yue Zhong , Jiawen Kang , Yongju Tong , Hong-Ning Dai , Dong In Kim , Abbas Jamalipour , Shengli Xie This is my paper · ORCID
classification cs.AI
keywords low-altitudeeconomyUAVnetworksLyapunovstabilitycontrol-communicationco-designStackelberggameresourcepricingdeepreinforcementlearningstructuredpruning
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper argues that in low-altitude UAV networks, communication resources should be allocated with an explicit eye on the physical stability of the control loops they serve, not just on throughput. It derives a closed-form mapping from a Lyapunov stability condition to a minimum bandwidth per user, then embeds that floor in a Stackelberg game where UAVs set bandwidth prices and users choose how much to request. To make the game tractable on battery-powered UAVs, the authors propose a structured-pruning version of Proximal Policy Optimization that shrinks the neural network during training. Simulations show the resulting prices keep every link above its stability bandwidth while raising system utility, and that pruning can improve final reward by acting as a regularizer.

What carries the argument

The load-bearing object is the Lyapunov descent inequality for the quadratic function V=ζ^T P ζ on the augmented state (position, velocity, previous control). With the packet-success model it gives the required success probability Γ; Markov's inequality converts Γ into a latency budget and then into the bandwidth floor κ_min. In the Stackelberg game, the follower's utility is strictly concave, yielding κ*_i=ϑ_i/ϱ_n−1/H_i,n; the leader's utility is concave too, so the optimal price is the unconstrained optimum clipped by a stability cap and a capacity floor. Dynamic structured pruning (binary masks with a cubic sparsity schedule) compresses the actor–critic network during training.

What would settle it

Configure a hardware-in-the-loop UAV tracking test, give every link exactly the derived κ_min, and count how often the Lyapunov function fails to decrease at the required rate; if the empirical violation frequency exceeds 1−Γ under the assumed packet-loss model, the closed-form mapping under-provisions bandwidth.

Watch

Extended reading notes

Core claim

The central claim is that UAV control stability can be converted, in closed form, into communication constraints. The paper models the UAV–user loop as a sensing–communication–computing–control system with fixed sampling, derives a Lyapunov descent condition for the augmented tracking state, and uses Markov's inequality to turn the required success probability into a latency budget. That budget yields a per-user bandwidth floor κ_min = S_i/(T_budget log2(1+SNR)). These floors are embedded in a Stackelberg game: UAVs lead by pricing bandwidth, users follow by requesting bandwidth, and backward induction gives a unique optimal price in one of three regimes—unconstrained, stability-capped, or c

Load-bearing premise

The whole mapping rests on treating the control loop as synchronous and linear, with latency appearing only as a shortened actuation window and packet drops as the only disturbance; if real delays cause missed, reordered, or asynchronous updates, the derived bandwidth floor is not a stability guarantee.

Editorial extensions

If this is right

  • Stability becomes a hard lower bound on allocated bandwidth, so control requirements can be checked at the resource allocation layer.
  • The optimal price is the unconstrained profit-maximizing price capped when stability is at risk and floored when capacity is at risk, making pricing a distributed stability safeguard.
  • Users with higher urgency weights receive larger bandwidth at the equilibrium, so the game inherently prioritizes control-critical links.
  • Dynamic structured pruning produces a smaller policy network with no loss—and sometimes a gain—in test reward, suggesting DRL controllers can run on energy-limited UAVs.
  • If the equilibrium is reached, no UAV or user can improve utility by changing price or request alone, giving a stable operating point for the closed loop.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the derivation only needs a monotone link between a resource and latency, the same stability-floor idea could be applied to transmit power, computation frequency, or schedule slots, not just bandwidth.
  • The Markov-inequality step is sufficient but not necessary, so the computed κ_min may be conservative; in deployments with well-characterized delay tails, a tighter probabilistic bound could lower the floor without endangering stability.
  • The pruning-as-regularization result suggests sparse policies might generalize better in other control-oriented DRL tasks, independent of the pricing game.
  • A natural stress test is to replace the fixed sampling period with a jittered or event-triggered sampler; the Lyapunov argument would need a different treatment, but the price-floor architecture could remain.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes a unified Sensing-Communication-Computing-Control (SC3) framework for UAV-assisted low-altitude economy networks. The authors use Lyapunov stability theory to derive a mapping from control-loop stability to per-user bandwidth constraints (Proposition 1), formulate the resource allocation problem as a Stackelberg game with UAVs as leaders and users as followers, and solve the game with a pruning-based PPO algorithm that dynamically prunes neural network structures during training. Simulation results are presented to show that the proposed scheme keeps the allocated bandwidth above the derived stability threshold and improves system utility compared to baselines.

Significance. If the stability-to-bandwidth mapping were rigorously established, the paper would offer a valuable cross-layer framework that ties physical control stability to communication resource allocation in UAV networks. The Stackelberg game formulation with a closed-form equilibrium and the lightweight pruning-based PPO are practical and relevant contributions for edge deployment. However, the central theoretical claim—that the proposed scheme 'secures control loop stability'—rests on Proposition 1, whose proof has load-bearing gaps: the state dependence of the success threshold is not handled, and the per-user bandwidth bound is at best a sufficient condition rather than the claimed minimum. The simulations also do not directly verify the Lyapunov descent condition. The framework is promising, but the current version does not adequately support its central stability guarantee.

major comments (4)
  1. [§III-B, Proposition 1 (Eqs. (19)–(22))] The success threshold Γ_n in Eq. (20) depends on the current control state ζ(k) through V_close and V_open, making D_req and T_budget state-dependent. The paper nevertheless treats κ_min_i as a fixed constraint in Problem 1 (23a) and in the Stackelberg price cap (28a). No mechanism is provided for resolving this state dependence (e.g., worst-case bound, state augmentation, or online estimation). Thus the derived bandwidth lower bound does not, by itself, enforce the state-dependent Lyapunov condition (17a).
  2. [§III-B, Proposition 1 (Eqs. (15), (19), (22))] The aggregate latency T_n in Eq. (15) is the maximum over connected users. The proof expands the Lyapunov expectation with a single-user success probability ξ_i,n and derives a per-user bound from E[T_n] ≤ D_req. For a maximum, satisfying E[T_i,n^total] ≤ D_req for each user does not imply E[max_i T_i,n^total] ≤ D_req; conversely, the per-user deterministic bound S_i/R_i ≤ T_budget is sufficient but not necessary. The claim that κ_min_i is the 'minimum bandwidth required' is therefore not established. The proposition should be restated as a sufficient condition and proved accordingly.
  3. [§III-B, Eq. (27)–(30)] The follower's best response κ*_i = ϑ_i/ϱ_n − 1/H_i,n can be negative when ϱ_n > ϑ_i H_i,n; the non-negativity constraint is not imposed. Additionally, the stability price cap in (28a) should be ϱ_n = min_i ϑ_i/(κ_min_i + 1/H_i,n) over all connected users, but the paper defines a single ϱ_n and drops the user index in (30). Without these corrections, the closed-form equilibrium may be infeasible and the stability constraint may be violated for some users.
  4. [§V, Fig. 2] The simulations verify that the allocated bandwidth exceeds the derived 'min requirement,' but they do not directly validate the Lyapunov descent condition (6) or the probabilistic stability guarantee. Since the central claim is securing control-loop stability, the paper should report a direct stability metric (e.g., empirical decay of V(ζ(k)), or empirical success probability P(T_n ≤ e_n) compared with the threshold Γ_n). Without such evidence, the stability claim is not empirically supported.
minor comments (5)
  1. [Fig. 1] Typo: 'Stakelberg' should be 'Stackelberg' in the figure caption.
  2. [§III-B, Eq. (28a)] The notation ϱ_n is used both for the price variable and for the upper bound; please clarify, and explicitly define the min over i for the stability cap.
  3. [Table I] The Lyapunov decay rate ρ_n = 0.95 is given without justification or sensitivity analysis; stability guarantees depend critically on this parameter.
  4. [§V, Fig. 4] The explanation of the non-monotonic effect of pruning start epoch is speculative; no error bars, confidence intervals, or multiple seeds are reported.
  5. [Algorithm 1] Line 22: 'Updatemask' should be 'Update mask'. Also, the pruning threshold in Eq. (36) is not clearly defined in terms of the importance score n_j^(h); please formalize.

Circularity Check

1 steps flagged · score 4.0 of 10

Stability 'confirmation' in Section V is tautological: the min-bandwidth threshold is imposed as constraint (23a), then its satisfaction is reported as evidence of control-loop stability.

  1. self definitional [Section V (Numerical Results), Fig. 2(d)/(h) analysis; cf. Eq. (23a) in Section III-B]
    "The 'min requirement' denotes the hard lower bound derived via Lyapunov optimization, which is necessary to guarantee the probabilistic stability of the physical control loop. It can be observed that the actual allocated bandwidth strictly supersedes this theoretical threshold for all users. This result confirms that the proposed strategy successfully enforces physical system stability as a baseline constraint"

    Problem 1 at Eq. (23a) imposes κ_i ≥ S_i/(T_budget log2(1+SNR_i,n)) as a hard constraint of the optimization, so any feasible (or converged) solution satisfies the 'min requirement' by construction. The paper reports this satisfaction as simulation evidence that the scheme 'successfully enforces physical system stability,' but no independent closed-loop state trajectory is shown; the observed inequality is the same inequality that was imposed in the problem definition. The stability conclusion therefore reduces to restating the constraint rather than testing it.

full rationale

The central derivation chain (Lyapunov condition (17a) -> success-probability threshold Γ -> D_req -> bandwidth bound (18) -> Stackelberg constraints (23a)) is not circular: (18) is a sufficiency argument, not a definitional identity with (17a). The proof contains a genuine correctness gap (Markov's inequality is applied to E[max_i T_i] after per-user latency terms are subtracted), but that is a mathematical error, not a circular reduction, and the reader's skepticism is better placed under correctness risk. The pruning-based PPO component is supported by self-citations [8], [40], but it is benchmarked against standard PPO in Figs. 3-6 within this paper, so the central claim does not reduce to a self-citation chain. The one circular element is the numerical 'stability confirmation' in Section V: the optimization already contains κ_i ≥ κ_i^min as a constraint (23a), so observing that allocated bandwidth exceeds that threshold cannot independently validate control-loop stability. This warrants a moderate circularity score, not a high one, because the Stackelberg game and pruning algorithm have independent content and the paper does provide external-looking comparisons against baselines.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

Abstract-only audit. The two constants named in the law (β_1, N) are the main free inputs; additional model parameters are implied by the 'local potential' but not visible. The axioms are the standard quantum-mechanics framework plus the domain assumption that a local potential captures all relevant physics.

free parameters (3)
  • β_1 (universal slope) = ≈1.44 as reported in the abstract
    The law d_{n,1}=β_1 n+N is determined by this constant; the abstract labels it universal, but it is a fitted output, not derived from QCD.
  • N (family-dependent intercept) = ≈2.49 as reported in the abstract
    The intercept absorbs the ground-state physics; it varies per family and is fixed to the data.
  • Local-potential parameters (potential strength, constituent quark masses)
    The abstract does not itemize them; from the abstract alone they are unverifiable inputs to the claimed derivation.
assumptions (3)
  • domain assumption Mesons are quark-antiquark bound states described by a local potential in a relativistic Schrödinger/Klein-Gordon-style equation.
    The entire derivation is claimed to follow from this model; the abstract introduces it without justification from QCD.
  • standard math The eigenvalue problem of that equation has a spectrum whose large-n behavior yields a linear law.
    The abstract claims an analytic derivation; the validity of the expansion or approximation used is not shown in the abstract.
  • domain assumption Observed and lattice masses can be associated unambiguously with radial quantum numbers n.
    Identification of excited meson states is known to be ambiguous in spectroscopy; the abstract gives no discussion of selection criteria.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Lyapunov Stability-Aware Stackelberg Game for Low-Altitude Economy: A Control-Oriented Pruning-Based DRL Approach." pith.science (2026). https://pith.science/paper/OYV6OBRL

@misc{pith2026260201131,
  author       = {Pith},
  title        = {Pith review of: Lyapunov Stability-Aware Stackelberg Game for Low-Altitude Economy: A Control-Oriented Pruning-Based DRL Approach},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OYV6OBRL}},
  note         = {Machine review of arXiv:2602.01131}
}
read the original abstract

With the rapid expansion of the low-altitude economy, Unmanned Aerial Vehicles (UAVs) serve as pivotal aerial base stations supporting diverse services from users, ranging from latency-sensitive critical missions to bandwidth-intensive data streaming. However, the efficacy of such heterogeneous networks is often compromised by the conflict between limited onboard resources and stringent stability requirements. Moving beyond traditional throughput-centric designs, we propose a Sensing-Communication-Computing-Control closed-loop framework that explicitly models the impact of communication latency on physical control stability. To guarantee mission reliability, we leverage the Lyapunov stability theory to derive an intrinsic mapping between the state evolution of the control system and communication constraints, transforming abstract stability requirements into quantifiable resource boundaries. Then, we formulate the resource allocation problem as a Stackelberg game, where UAVs (as leaders) dynamically price resources to balance load and ensure stability, while users (as followers) optimize requests based on service urgency. Furthermore, addressing the prohibitive computational overhead of standard Deep Reinforcement Learning (DRL) on energy-constrained edge platforms, we propose a novel and lightweight pruning-based Proximal Policy Optimization (PPO) algorithm. By integrating a dynamic structured pruning mechanism, the proposed algorithm significantly compresses the neural network scale during training, enabling the UAV to rapidly approximate the game equilibrium with minimal inference latency. Simulation results demonstrate that the proposed scheme effectively secures control loop stability while maximizing system utility in dynamic low-altitude environments.

Figures

Figures reproduced from arXiv: 2602.01131 by the authors.

Figure 1
Figure 1. Illustration of the integrated SC3 -based and Lyapunov stability-aware Stackelberg solved by lightweight pruning-based MADRL for the low-altitude economy framework. The framework illustrates the integration of heterogeneous user demands, the Lyapunov-driven fusion of SC3 dynamics, and the Stackelberg game formulation, ultimately solved by the dynamic structured pruning-based PPO algorithm. largely overlooking the un… view at source ↗
Figure 2
Figure 2. Comparative analysis of network topology and bandwidth allocation strategies under different load conditions. The top row (a)-(d) presents the scenario [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. Test reward comparison of the proposed pruning-based PPO algorithm [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Test reward comparison of the proposed pruning-based PPO algorithm [PITH_FULL_IMAGE:figures/full_fig_p010_4.png]
Figure 6
Figure 6. Figure 6: Test reward comparison of the proposed pruning-based PPO algorithm [PITH_FULL_IMAGE:figures/full_fig_p011_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 2 linked inside Pith

  1. [1]

    Coded caching for reli- able map dissemination in symbiotic communication aided emergency UA V systems,

    Z. Tian, L. Wang, L. Xu, C. Xu, and A. Fei, “Coded caching for reli- able map dissemination in symbiotic communication aided emergency UA V systems,”IEEE Transactions on Cognitive Communications and Networking, vol. 10, no. 5, pp. 1663–1677, 2024

    2024

  2. [2]

    Survey on UA V cellular communications: Practical aspects, standardization advancements, regulation, and security challenges,

    A. Fotouhi, H. Qiang, M. Ding, M. Hassan, L. G. Giordano, A. Garcia- Rodriguez, and J. Yuan, “Survey on UA V cellular communications: Practical aspects, standardization advancements, regulation, and security challenges,”IEEE Communications surveys & tutorials, vol. 21, no. 4, pp. 3417–3442, 2019

    2019

  3. [3]

    Population- invariant MADRL for AoI-aware UA V trajectory design and communi- cation scheduling in wireless sensor networks,

    X. Zhou, J. Xiong, H. Zhao, C. Yan, H. Wang, and J. Wei, “Population- invariant MADRL for AoI-aware UA V trajectory design and communi- cation scheduling in wireless sensor networks,”IEEE Internet of Things Journal, vol. 12, no. 3, pp. 2545–2561, 2025

    2025

  4. [4]

    Advancing the control of low-altitude wireless networks: architecture, design principles, and future directions,

    H. Jin, W. Yuan, J. Wu, J. Wang, D. Niyato, X. Wang, G. K. Kara- giannidis, Z. Lin, Y . Gong, D. I. Kimet al., “Advancing the control of low-altitude wireless networks: architecture, design principles, and future directions,”npj Wireless Technology, vol. 2, no. 1, p. 2, 2026

    2026

  5. [5]

    Joint task offloading and resource allocation in aerial-terrestrial UA V networks with edge and fog computing for post-disaster rescue,

    G. Sun, L. He, Z. Sun, Q. Wu, S. Liang, J. Li, D. Niyato, and V . C. Leung, “Joint task offloading and resource allocation in aerial-terrestrial UA V networks with edge and fog computing for post-disaster rescue,” IEEE Transactions on Mobile Computing, vol. 23, no. 9, pp. 8582–8600, 2024

    2024

  6. [6]

    Co-design of sensing, communications, and control for low-altitude wireless networks,

    H. Jin, J. Wu, W. Yuan, F. Liu, and Y . Cui, “Co-design of sensing, communications, and control for low-altitude wireless networks,”arXiv preprint arXiv:2506.20970, 2025

    arXiv 2025

  7. [7]

    On the AoI- aware status update in buffer-aided wireless-powered internet of things network,

    T. Wang, S. Wang, X. Lan, Y . Liu, Q. Chen, and P. Xiao, “On the AoI- aware status update in buffer-aided wireless-powered internet of things network,”IEEE Internet of Things Journal, vol. 11, no. 7, pp. 12 551– 12 566, 2024

    2024

  8. [8]

    Tiny multiagent DRL for twins migration in UA V metaverses: A multileader multifollower stackelberg game approach,

    J. Kang, Y . Zhong, M. Xu, J. Nie, J. Wen, H. Du, D. Ye, X. Huang, D. Niyato, and S. Xie, “Tiny multiagent DRL for twins migration in UA V metaverses: A multileader multifollower stackelberg game approach,” IEEE Internet of Things Journal, vol. 11, no. 12, pp. 21 021–21 036, 2024. 12

    2024

Show all 44 references
  1. [9]

    Optimizing dynamic spectrum sharing in UA V-assisted networks: Hybrid two-stage stackel- berg game approach,

    Q. Wang, Z. Li, M. Chen, H. Zhao, and H. Zhu, “Optimizing dynamic spectrum sharing in UA V-assisted networks: Hybrid two-stage stackel- berg game approach,”IEEE Internet of Things Journal, vol. 12, no. 19, pp. 40 517–40 530, 2025

    2025

  2. [10]

    Multi-attribute auction-based resource allocation for twins migration in vehicular metaverses: A GPT-based DRL approach,

    Y . Tong, J. Chen, M. Xu, J. Kang, Z. Xiong, D. Niyato, C. Yuen, and Z. Han, “Multi-attribute auction-based resource allocation for twins migration in vehicular metaverses: A GPT-based DRL approach,”IEEE Transactions on Cognitive Communications and Networking, vol. 11, no. 1, ...

    2025

  3. [11]

    Neural combinatorial optimization with reinforcement learning in industrial engineering: a survey,

    K. Chung, C. Lee, and Y . Tsang, “Neural combinatorial optimization with reinforcement learning in industrial engineering: a survey,”Artifi- cial Intelligence Review, vol. 58, no. 5, p. 130, 2025

    2025

  4. [12]

    Integrated sensing and communications for low-altitude economy: A deep reinforcement learning approach,

    X. Ye, Y . Mao, X. Yu, S. Sun, L. Fu, and J. Xu, “Integrated sensing and communications for low-altitude economy: A deep reinforcement learning approach,”IEEE Transactions on Wireless Communications, vol. 25, pp. 351–367, 2026

    2026

  5. [13]

    MADQN-enhanced computation offloading and resource allocation for 6G low-altitude economy vehicular networks,

    B. Hu, H. Liu, J. Du, M. L ´opez-Ben´ıtez, C. Wu, X. Chu, and D. Niyato, “MADQN-enhanced computation offloading and resource allocation for 6G low-altitude economy vehicular networks,”IEEE Transactions on Cognitive Communications and Networking, vol. 12, pp. 2603–2617, 2026

    2026

  6. [14]

    A two phases multiobjective trajectory optimization scheme for multi-UGVs in the sight of the first aid scenario,

    R. Chai, K. Chen, B. Hua, Y . Lu, Y . Xia, X.-M. Sun, G.-P. Liu, and W. Liang, “A two phases multiobjective trajectory optimization scheme for multi-UGVs in the sight of the first aid scenario,”IEEE Transactions on Cybernetics, vol. 54, no. 9, pp. 5078–5091, 2024

    2024

  7. [15]

    An accurate UA V 3-D path planning method for disaster emergency response based on an improved multiobjective swarm intelligence algorithm,

    Y . Wan, Y . Zhong, A. Ma, and L. Zhang, “An accurate UA V 3-D path planning method for disaster emergency response based on an improved multiobjective swarm intelligence algorithm,”IEEE Transactions on Cybernetics, vol. 53, no. 4, pp. 2658–2671, 2023

    2023

  8. [16]

    AoI-aware incentive mechanism for UA V-assisted mobile crowdsensing: A contract-theoretic approach,

    Y . Guo, Y . Chen, H. Li, Y . Wu, and J. Huang, “AoI-aware incentive mechanism for UA V-assisted mobile crowdsensing: A contract-theoretic approach,”IEEE Transactions on Mobile Computing, vol. 25, no. 2, pp. 1660–1677, 2026

    2026

  9. [17]

    Optimizing spectrum sharing in UA V swarms: A stackelberg game-based incentive mechanism,

    Q. Wang, Y . Shen, L. Xu, H. Zhang, H. Zhao, and H. Zhu, “Optimizing spectrum sharing in UA V swarms: A stackelberg game-based incentive mechanism,”IEEE Transactions on Vehicular Technology, vol. 75, no. 1, pp. 1443–1454, 2026

    2026

  10. [18]

    Joint task offloading and resource allocation in aerial-terrestrial UA V networks with edge and fog computing for post-disaster rescue,

    G. Sun, L. He, Z. Sun, Q. Wu, S. Liang, J. Li, D. Niyato, and V . C. M. Leung, “Joint task offloading and resource allocation in aerial-terrestrial UA V networks with edge and fog computing for post-disaster rescue,” IEEE Transactions on Mobile Computing, vol. 23, no. 9, pp. 8...

    2024

  11. [19]

    Generative AI for lyapunov optimization theory in UA V-based low- altitude economy networking,

    Z. Liu, D. Niyato, J. Wang, G. Sun, L. Huang, Z. Gao, and X. Wang, “Generative AI for lyapunov optimization theory in UA V-based low- altitude economy networking,”IEEE Network, pp. 1–9, 2026

    2026

  12. [20]

    Pruning-based deep reinforcement learning for task offloading in end-edge-cloud collabo- rative mobile edge computing,

    H. Yang, H. Zhang, F. Luo, F. Liu, and H. Chen, “Pruning-based deep reinforcement learning for task offloading in end-edge-cloud collabo- rative mobile edge computing,”Journal of Computing and Electronic Information Management, vol. 13, no. 1, pp. 1–9, 2024

    2024

  13. [21]

    Compact learning model for dynamic off-chain routing in blockchain-based IoT,

    Z. Li, W. Su, M. Xu, R. Yu, D. Niyato, and S. Xie, “Compact learning model for dynamic off-chain routing in blockchain-based IoT,”IEEE Journal on Selected Areas in Communications, vol. 40, no. 12, pp. 3615– 3630, 2022

    2022

  14. [22]

    Non-isotropy regularization for proxy-based deep metric learning,

    K. Roth, O. Vinyals, and Z. Akata, “Non-isotropy regularization for proxy-based deep metric learning,” inProceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2022, pp. 7420– 7430

    2022

  15. [23]

    Solutions for sus- tainable and resilient communication infrastructure in disaster relief and management scenarios,

    B. Karaman, I. Basturk, S. Taskin, E. Zeydan, F. Kara, E. A. Beyazıt, M. Camelo, E. Bj ¨ornson, and H. Yanikomeroglu, “Solutions for sus- tainable and resilient communication infrastructure in disaster relief and management scenarios,”IEEE Communications Surveys & Tutorials, v...

    2026

  16. [24]

    Emerging UA V technology for disaster detection, mitigation, response, and preparedness,

    A. Khan, S. Gupta, and S. K. Gupta, “Emerging UA V technology for disaster detection, mitigation, response, and preparedness,”Journal of Field Robotics, vol. 39, no. 6, pp. 905–955, 2022

    2022

  17. [25]

    Optimizing resource allocation in URLLC for real-time wireless control systems,

    B. Chang, L. Zhang, L. Li, G. Zhao, and Z. Chen, “Optimizing resource allocation in URLLC for real-time wireless control systems,”IEEE Transactions on Vehicular Technology, vol. 68, no. 9, pp. 8916–8927, 2019

    2019

  18. [26]

    Zero-order hold discretization of general state space systems with input delay,

    G. Pechlivanidou and N. Karampetakis, “Zero-order hold discretization of general state space systems with input delay,”IMA Journal of Mathematical Control and Information, vol. 39, no. 2, pp. 708–730, 2022

    2022

  19. [27]

    Wireless network design for control systems: A survey,

    P. Park, S. Coleri Ergen, C. Fischione, C. Lu, and K. H. Johansson, “Wireless network design for control systems: A survey,”IEEE Com- munications Surveys & Tutorials, vol. 20, no. 2, pp. 978–1013, 2018

    2018

  20. [28]

    Dynamic gain adaptation in linear quadratic regulators,

    A. Komaee, “Dynamic gain adaptation in linear quadratic regulators,” IEEE Transactions on Automatic Control, vol. 69, no. 8, pp. 5094–5108, 2023

    2023

  21. [29]

    Lyapunov-guided deep reinforcement learning for semantic-aware AoI minimization in UA V-assisted wireless networks,

    Y . Long, S. Gong, S. Sun, G. C. F. Lee, L. Li, and D. Niyato, “Lyapunov-guided deep reinforcement learning for semantic-aware AoI minimization in UA V-assisted wireless networks,”IEEE Transactions on Wireless Communications, vol. 24, no. 8, pp. 6351–6364, 2025

    2025

  22. [30]

    SGCS: A cost-effective quality control system for strategic workers in mobile crowd sensing,

    H. Wang, A. Liu, and N. N. Xiong, “SGCS: A cost-effective quality control system for strategic workers in mobile crowd sensing,”IEEE Transactions on Network Science and Engineering, vol. 12, no. 2, pp. 1146–1158, 2025

    2025

  23. [31]

    LEO-satellite-assisted UA V path optimiza- tion for space–air–ground internet of remote things networks,

    X. Liu, L. Lv, and Q. Yang, “LEO-satellite-assisted UA V path optimiza- tion for space–air–ground internet of remote things networks,”Space: Science & Technology, vol. 5, p. 0280, 2025

    2025

  24. [32]

    Generative semantic communication via textual prompts: Latency performance tradeoffs,

    M. Ren, L. Qiao, L. Yang, Z. Gao, J. Chen, M. B. Mashhadi, P. Xiao, R. Tafazolli, and M. Bennis, “Generative semantic communication via textual prompts: Latency performance tradeoffs,”IEEE Transactions on Vehicular Technology, vol. 74, no. 9, pp. 14 843–14 848, 2025

    2025

  25. [33]

    Joint optimization of completion ratio and latency of offloaded tasks with multiple priority levels in 5G edge,

    P. Fard Moshiri, M. Simsek, and B. Kantarci, “Joint optimization of completion ratio and latency of offloaded tasks with multiple priority levels in 5G edge,”IEEE Transactions on Network and Service Man- agement, vol. 22, no. 2, pp. 1357–1371, 2025

    2025

  26. [34]

    Ogata,Modern control engineering

    K. Ogata,Modern control engineering. Prentice hall, 2010

    2010

  27. [35]

    Fixed-time collision-free fault- tolerant formation control of multi-UA Vs under actuator faults,

    Q. Miao, K. Zhang, and B. Jiang, “Fixed-time collision-free fault- tolerant formation control of multi-UA Vs under actuator faults,”IEEE Transactions on Cybernetics, vol. 54, no. 6, pp. 3679–3691, 2024

    2024

  28. [36]

    S. M. Ross,Introduction to probability and statistics for engineers and scientists. Academic press, 2020

    2020

  29. [37]

    Optimizing resource allocation for multi-modal semantic communication in mobile AIGC networks: A diffusion-based game approach,

    J. Liu, M. Xiao, J. Wen, J. Kang, R. Zhang, T. Zhang, D. Niyato, W. Zhang, and Y . Liu, “Optimizing resource allocation for multi-modal semantic communication in mobile AIGC networks: A diffusion-based game approach,”IEEE Transactions on Cognitive Communications and Networking...

    2025

  30. [38]

    Blockchain-assisted twin migration for vehicular metaverses: A game theory approach,

    Y . Zhong, J. Wen, J. Zhang, J. Kang, Y . Jiang, Y . Zhang, Y . Cheng, and Y . Tong, “Blockchain-assisted twin migration for vehicular metaverses: A game theory approach,”Transactions on Emerging Telecommunica- tions Technologies, vol. 34, no. 12, p. e4856, 2023

    2023

  31. [39]

    From black-box to white-box: Interpretable deep reinforcement learning with kolmogorov-arnold networks for autonomous driving,

    J. Huang, R. Zhou, M. Li, H. Li, Y . Liu, and X. Song, “From black-box to white-box: Interpretable deep reinforcement learning with kolmogorov-arnold networks for autonomous driving,”Transportation Research Part C: Emerging Technologies, vol. 182, p. 105386, 2026

    2026

  32. [40]

    Compressing deep reinforcement learning networks with a dynamic structured prun- ing method for autonomous driving,

    W. Su, Z. Li, M. Xu, J. Kang, D. Niyato, and S. Xie, “Compressing deep reinforcement learning networks with a dynamic structured prun- ing method for autonomous driving,”IEEE Transactions on Vehicular Technology, vol. 73, no. 12, pp. 18 017–18 030, 2024

    2024

  33. [41]

    Hadamard product in deep learning: Introduction, advances and challenges,

    G. G. Chrysos, Y . Wu, R. Pascanu, P. H. Torr, and V . Cevher, “Hadamard product in deep learning: Introduction, advances and challenges,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 47, no. 8, pp. 6531–6549, 2025

    2025

  34. [42]

    Bi-LSTM based multi-agent DRL with computation-aware pruning for agent twins migration in vehicular embodied AI networks,

    Y . Wei, Z. Zeng, Y . Zhong, J. Kang, R. W. Liu, and M. S. Hossain, “Bi-LSTM based multi-agent DRL with computation-aware pruning for agent twins migration in vehicular embodied AI networks,”arXiv preprint arXiv:2505.06378, 2025

    2025 arXiv

  35. [43]

    Sustainable diffusion-based incentive mechanism for generative AI-driven digital twins in industrial cyber-physical systems,

    J. Wen, J. Kang, D. Niyato, Y . Zhang, and S. Mao, “Sustainable diffusion-based incentive mechanism for generative AI-driven digital twins in industrial cyber-physical systems,”IEEE Transactions on In- dustrial Cyber-Physical Systems, vol. 3, pp. 139–149, 2025

    2025

  36. [44]

    Age-of-task-aware AA V-based mobile edge computing techniques in emergency rescue applications,

    X. Peng, X. Lan, and Q. Chen, “Age-of-task-aware AA V-based mobile edge computing techniques in emergency rescue applications,”IEEE Internet of Things Journal, vol. 12, no. 7, pp. 8909–8930, 2025

    2025

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.