Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Bi-LSTM based Multi-Agent DRL with Computation-aware Pruning for Agent Twins Migration in Vehicular Embodied AI Networks

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read In vehicular embodied AI networks, offloading an AI agent to a roadside unit has a unique Stackelberg equilibrium when the task's importance exceeds the product of bandwidth price and delay tolerance, and a Bi-LSTM multi-agent learner…

desk verdict The paper's central equilibrium proof is wrong—Eq. (10) misdifferentiates the follower utility—so the claimed unique Stackelberg equilibrium does not follow; the DRL and pruning pieces are standard and the experiments are not independently verifiable. read the letter →

arxiv 2505.06378 v1 pith:UZAWIMJT submitted 2025-05-09 cs.GT cs.AI

Yuxiang Wei , Zhuoqi Zeng , Yue Zhong , Jiawen Kang , Ryan Wen Liu , M. Shamim Hossain This is my paper · ORCID
classification cs.GTcs.AI MSC 91A65
keywords vehicularembodiedAInetworkstwinmigrationStackelberggamemulti-agentdeepreinforcementlearningbidirectionalLSTMnetworkpruningPatheXclusionbandwidthallocation
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

The paper argues that migrating vehicular embodied AI agents to roadside units can be treated as a multi-leader multi-follower Stackelberg game in which roadside units set bandwidth prices and vehicles respond with bandwidth requests. Its main theoretical claim is that this game has a unique Stackelberg equilibrium once each task's importance is high enough relative to the price and the vehicle's delay tolerance. On top of that equilibrium analysis the paper builds a multi-agent reinforcement learning algorithm using bidirectional LSTMs, and a computation-aware pruning step that shrinks each vehicle's actor network according to its onboard computing power. The aim is to keep roadside-unit workloads balanced while keeping migration latency low, and the reported simulations on embedded hardware show the pruned learner outperforming baselines in cumulative reward.

What carries the argument

The central object is the MLMF Stackelberg game with QoS-weighted utilities: AV utility is a logarithmic Weber–Fechner revenue term minus bandwidth cost, and RSU utility is bandwidth revenue minus base cost. The argument is carried by the closed-form follower best response and by a standard-function proof for the leader best response, which together establish the unique equilibrium condition $\alpha_v \ge e p_r T^{\max}_v$. To reach that equilibrium in a partially observable setting, the algorithm uses Bi-LSTM actors that ingest the previous $L$ rounds of prices and demands, a PPO-style clipped surrogate loss, and a Path eXclusion pruning step that scores weights by a path-activation by path-weights product so the pruned subnetworks keep the task-critical pathways.

What would settle it

Run TMABLPPO in the described environment, record the converged bandwidth requests and RSU prices, and compare them with the analytic best responses $b^\star_{rv} = 1/p_r - e T^{\max}_v/\alpha_v$ and the leader's first-order condition across the regime $\alpha_v \ge e p_r T^{\max}_v$; a systematic gap between learned and analytic actions would refute the claim that the algorithm approximates the equilibrium.

Watch

Extended reading notes

Core claim

Under the condition $\alpha_v \ge e p_r T^{\max}_v$, the paper derives a unique Stackelberg equilibrium $(\mathbf{B}^\star,\mathbf{P}^\star)$ for the AV–RSU bandwidth market. The follower-level best response is closed form: vehicle $v$ requests $b^\star_{rv} = 1/p_r - e T^{\max}_v/\alpha_v$ from RSU $r$ when the condition holds, and requests zero bandwidth otherwise. The leader-level best response is shown to be a standard function, which implies a unique Nash equilibrium among RSUs and therefore a unique Stackelberg equilibrium. The TMABLPPO algorithm is then proposed to approximate this equilibrium through decentralized coordination, and the Path eXclusion pruning method compresses the actor networks so that vehicles with different computation tiers execute the learned policies within latency constraints.

Load-bearing premise

The load-bearing premise is that the reward the simulator gives to the DRL agents equals the AV and RSU utility functions used in the equilibrium analysis; the paper never states that reward, so the connection between the learned policies and the claimed Stackelberg equilibrium rests on this unstated matching.

Editorial extensions

If this is right

  • Roadside units can use the equilibrium condition as an admission rule: vehicles whose task importance falls below $e p_r T^{\max}_v$ should buy no bandwidth from that RSU, which offloads the cheapest traffic automatically.
  • The closed-form best response gives a direct benchmark for any learned policy; a DRL agent that does not converge to $b^\star_{rv}$ in the stated regime is not solving the game as modeled.
  • Bi-LSTM temporal context shortens training to reach optimal strategies, so the same reward signal yields faster convergence than one-step observation baselines.
  • Pruning at 90% density preserves or improves cumulative reward, so the latency gain from compression does not cost task performance in the tested regime.
  • At 33% density the model keeps about 68% of its original reward, indicating graceful degradation under aggressive compression.

Reading between the lines

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

  • The analytic equilibrium could be used directly as a reward-shaping term or a curriculum prior for the DRL training, which would make the claimed approximation testable independently of the simulation's reward design.
  • The same leader–follower structure applies to other digital-twin migration markets, such as UAV or metaverse service migration, so the uniqueness theorem transfers wherever utility has the same log-revenue-minus-cost form.
  • One could replace the fixed density tiers with an online latency–accuracy Pareto search per vehicle, since the paper's three-tier thresholds are a heuristic discretization of a continuous trade-off.
  • The PX saliency's reliance on the neural tangent kernel suggests a principled connection between pruning and trainability: weights on high-curvature paths are kept, which may explain why 90%-density pruning does not degrade reward.
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

3 major / 5 minor

Summary. This paper models the migration of vehicular embodied AI agent twins (VEAATs) from autonomous vehicles to roadside units as a multi-leader multi-follower Stackelberg game over bandwidth pricing and bandwidth demand. The authors propose a Bi-LSTM multi-agent PPO algorithm (TMABLPPO) with an NTK-based Path eXclusion pruning scheme, claim a unique Stackelberg equilibrium under α_v ≥ e p_r T_max^v, and report simulation results showing improved rewards and RSU/AV utilities. The central theoretical claim is the existence of this equilibrium and the assertion that the DRL algorithm approximates it.

Significance. Should the equilibrium derivation and the reward design be correct, the paper would offer a useful integration of game-theoretic resource allocation with computation-aware DRL compression for edge-assisted vehicular AI migration. The system model is topical, the use of QoS-aware logarithmic utilities is reasonable, and the experiments are conducted on an embedded platform with a pruning method that is at least concretely specified. However, the mathematical analysis establishing the Stackelberg equilibrium contains a derivative error that propagates through the leader-level proof, and the DRL reward is never defined. These issues are load-bearing: without a correct best-response derivation and a reward that matches the utilities, the claimed approximation of the Stackelberg equilibrium is not established. The paper's main contribution is therefore conditional and, as written, not supportable.

major comments (3)
  1. [Section IV-A, Eqs. (10) and (12)] Eq. (10) misdifferentiates the follower utility in Eq. (5). With θ_rv = (1/p_r) / Σ_l (1/p_l), the derivative of U_F_v with respect to b_rv is Σ_r [α_v β θ_rv (α_v/T_max_v) / (e + α_v b_rv/T_max_v) − p_r], but Eq. (10) applies the prefactor α_v β θ_rv to both the logarithmic term and the −p_r term. The resulting first-order condition therefore cannot yield Eq. (12); the correct best response would be b_rv = α_v β θ_rv / p_r − e T_max_v / α_v, not b̂_rv = 1/p_r − e T_max_v / α_v. Because Eq. (12) is substituted into Eq. (14) and used in the proof of Theorem 1, the claimed uniqueness of the Stackelberg equilibrium does not follow.
  2. [Section IV-B, Eq. (17)] Eq. (17) is not the solution of the leader's aggregate first-order condition. Eq. (15) sets ∂U_L^r/∂y_r = Σ_v [ ... ] = 0, which requires solving a sum over all followers; Eq. (17) instead writes ŷ_j as a sum over v of the individual quadratic solutions. These operations are not equivalent unless the cross-terms vanish, and no justification is given. Consequently, the leader best-response function G(Y), the standard-function argument, and the claimed unique Nash equilibrium of the leader-level game are not established.
  3. [Section V-B and Algorithm 1] The reward function for the MADRL agents is never defined. Section V-B only states that RSUs and AVs generate 'rewards that combine immediate operational outcomes and long-term sustainability metrics,' and Algorithm 1 invokes R_t^r and R_t^v without equations. The paper's claim that TMABLPPO approximates the Stackelberg equilibrium requires the training reward to coincide with U_F and U_L from Eqs. (5) and (6), or to be a proxy whose optimum provably coincides with the equilibrium. With no reward specification, the numerical comparisons in Section VI demonstrate only that one DRL variant obtains a higher total of an unstated objective; they cannot validate equilibrium approximation.
minor comments (5)
  1. [Section III-C1, Eq. (5)] The logarithmic term in Eq. (5) uses α_r b_rv, while the task importance parameter is defined as α_v; the subscript appears to be a typo and should be α_v (or α_r must be defined).
  2. [Section III-C2, Eq. (7)] The latency constraint Σ_{j∈R} T_rv ≤ T_max^i mixes the indices r, v, and i; it should be a per-pair constraint such as T_rv ≤ T_max^v, and the sum over RSUs is unclear because T_rv already depends on r.
  3. [Section IV-B, Eq. (18)] The best response in Eq. (18) includes a zero branch, yet the subsequent standard-function proof asserts positivity of G(Y); the paper should clarify the domain on which positivity holds and correct the strict/weak inequality in the threshold condition.
  4. [Section V-C, Eq. (28)] The saliency expression ∂R/(∂θ^2) ⊙ θ^2 is ill-formed; it should presumably be (∂R/∂θ)^2 ⊙ θ^2. The dimension statement for the NTK in Eq. (29), written as an NK×NK matrix, is also confusing and should be clarified.
  5. [Section VI] The numerical section omits key experimental parameters (numbers of AVs and RSUs, α_v, β, c_r, p_max, and the simulator's transition dynamics) and reports no error bars or multiple seeds, which limits reproducibility and makes it difficult to interpret the reward comparisons in Figs. 3–8.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the equilibrium derivation and the learning/pruning components do not reduce to their own inputs; any defects are correctness or validation issues, not circular reasoning.

full rationale

I walked the claimed derivation chain in the paper. The Stackelberg equilibrium analysis in Section IV is attempted self-contained: the follower best response in Eq. (12) is obtained from the first-order condition of the paper's own utility function Eq. (5), and the leader best response in Eqs. (15)-(18) is obtained from the paper's own leader utility Eq. (14), with uniqueness imported from the standard-function argument in [37], an external reference. The self-citations in the paper, e.g., [8], [11], [12], and [26], are used for context, related work, or design inspiration; they are not the load-bearing justification for the central equilibrium claim. The DRL and pruning sections use standard PPO, Bi-LSTM, and Path eXclusion methodology, and the claim that the learned policies approximate the Stackelberg equilibrium is under-specified because the reward function is not explicitly equated to the utility functions U_F and U_L. However, that is a validation gap, not a circular reduction: the paper does not define the DRL objective in terms of the equilibrium in a way that would make the approximation true by construction. The pruning saliency in Eq. (35) is computed from the trained network's own NTK, but this is a self-referential importance measure for compression, not a prediction claimed to be equivalent to its inputs. Any algebraic error in Eqs. (10)-(12) or (17) would be a mathematical correctness issue, not a circularity issue, because the claimed result would still not be assumed as an input. Since no specific reduction of a claimed result to its own inputs can be exhibited, the appropriate circularity score is 0.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

No new physical entities, forces, or conserved quantities are introduced. The VEAAT concept is a software abstraction defined in prior work (ref [8]) and reused here.

free parameters (5)
  • beta (service perception marginal effect)
    Introduced in Eq. (4) and used in U_F; its value is never specified. The closed-form best response Eq. (12) implicitly assumes beta=1, so the derivation is inconsistent with the general model.
  • alpha_v (task importance)
    Per-AV task importance used in Eqs. (3)-(5) and in the positivity condition of Theorem 1; the paper does not specify how alpha_v is set or sampled in the experiments.
  • c_r (RSU base cost)
    Base cost in RSU utility Eq. (6); value not specified, but it appears in the leader's best response and the standard-function proof.
  • Pruning density thresholds = 80-100%, 40-80%, 0-40%
    Table I maps AV computation capability tiers to density intervals; these thresholds are chosen by hand with no derivation.
  • p_max (maximum bandwidth price)
    Constraint in leader problem P2 and action bound; value not specified.
assumptions (6)
  • domain assumption AVs and RSUs are rational utility maximizers with common knowledge of the game.
    The Stackelberg equilibrium and the DRL objective assume each agent maximizes its own utility without altruism or bounded rationality (Section III-C).
  • domain assumption The Weber-Fechner logarithmic law captures QoS perception.
    Eq. (4) assumes QoS revenue is logarithmic in bandwidth and task importance, with no empirical validation in this paper.
  • standard math Concavity (negative second derivative) implies a global maximum for the follower's utility.
    Used in Section IV-A to justify setting the first-order derivative to zero.
  • standard math A leader-level best response satisfying the standard function properties yields a unique Nash equilibrium.
    Invoked from ref [37] in the proof of Theorem 1.
  • domain assumption The Neural Tangent Kernel factorization in PX pruning is valid for the RL actor network.
    Eqs. (28)-(35) rely on NTK theory as applied to the trained actor, carried over from ref [16] and ref [38].
  • domain assumption The channel and latency model (Eqs. (1)-(2)) accurately represent the communication system.
    The simulation uses the Shannon rate and OTFS channel estimation formulas as ground truth.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bi-LSTM based Multi-Agent DRL with Computation-aware Pruning for Agent Twins Migration in Vehicular Embodied AI Networks." pith.science (2026). https://pith.science/paper/UZAWIMJT

@misc{pith2026250506378,
  author       = {Pith},
  title        = {Pith review of: Bi-LSTM based Multi-Agent DRL with Computation-aware Pruning for Agent Twins Migration in Vehicular Embodied AI Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UZAWIMJT}},
  note         = {Machine review of arXiv:2505.06378}
}
read the original abstract

With the advancement of large language models and embodied Artificial Intelligence (AI) in the intelligent transportation scenarios, the combination of them in intelligent transportation spawns the Vehicular Embodied AI Network (VEANs). In VEANs, Autonomous Vehicles (AVs) are typical agents whose local advanced AI applications are defined as vehicular embodied AI agents, enabling capabilities such as environment perception and multi-agent collaboration. Due to computation latency and resource constraints, the local AI applications and services running on vehicular embodied AI agents need to be migrated, and subsequently referred to as vehicular embodied AI agent twins, which drive the advancement of vehicular embodied AI networks to offload intensive tasks to Roadside Units (RSUs), mitigating latency problems while maintaining service quality. Recognizing workload imbalance among RSUs in traditional approaches, we model AV-RSU interactions as a Stackelberg game to optimize bandwidth resource allocation for efficient migration. A Tiny Multi-Agent Bidirectional LSTM Proximal Policy Optimization (TMABLPPO) algorithm is designed to approximate the Stackelberg equilibrium through decentralized coordination. Furthermore, a personalized neural network pruning algorithm based on Path eXclusion (PX) dynamically adapts to heterogeneous AV computation capabilities by identifying task-critical parameters in trained models, reducing model complexity with less performance degradation. Experimental validation confirms the algorithm's effectiveness in balancing system load and minimizing delays, demonstrating significant improvements in vehicular embodied AI agent deployment.

Figures

Figures reproduced from arXiv: 2505.06378 by the authors.

Figure 1
Figure 1. The system model for VEAAT migration. al. [25] utilized the Stackelberg framework to optimize re￾source allocation in the collaborative intelligent transportation systems. The above works highlight the resilience of Stackel￾berg game theory for hierarchical decision-making, especially for dealing with constrained resources and multiple players. With the continued progress in computing, networking, and the Internet o… view at source ↗
Figure 2
Figure 2. TMABLPPO algorithm’s Framework for the VEAAT migration. [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 5
Figure 5. Comparison of the total reward curves of [PITH_FULL_IMAGE:figures/full_fig_p010_5.png] view at source ↗
Figures from the paper (2 more)
Figure 6
Figure 6. Figure 6: Relation between the total reward and density in the trivial pruning rate threshold [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 9
Figure 9. Figure 9: Average utility of the RSUs with MABLPPO and MAA3C algorithms. [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

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

    cs.AI 2026-02 unverdicted novelty 6.0 of 10

    Radial excitations of ρ, π, φ, D, D_s, B, B_s, ψ and Υ mesons obey d_{n,1}=β_1 n + N (universal slope, family intercept), claimed here to follow analytically from a local potential.

Reference graph

Works this paper leans on

40 extracted references · 29 canonical work pages · cited by 1 Pith paper

  1. [1]

    Multiply: A multisensory object-centric embodied large language model in 3D world,

    Y . Hong, Z. Zheng, P. Chen, Y . Wang, J. Li, and C. Gan, “Multiply: A multisensory object-centric embodied large language model in 3D world,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , June 2024, pp. 26 406–26 416

    work page 2024

  2. [2]

    Embodied AI with large language models: A survey and new hri framework,

    M.-Y . Lin, O.-W. Lee, and C.-Y . Lu, “Embodied AI with large language models: A survey and new hri framework,” in 2024 International Conference on Advanced Robotics and Mechatronics (ICARM) . IEEE, 2024, pp. 978–983

    work page 2024

  3. [3]

    Language models meet world models: Embodied experiences enhance language models,

    J. Xiang, T. Tao, Y . Gu, T. Shu, Z. Wang, Z. Yang, and Z. Hu, “Language models meet world models: Embodied experiences enhance language models,” in Advances in Neural Information Processing Systems, A. Oh, T. Naumann, A. Globerson, K. Saenko, M. Hardt, and S. Levine, Eds., vol. 36. Curran Associates, Inc., 2023, pp. 75 392–75 412

    work page 2023

  4. [4]

    Llm-planner: Few-shot grounded planning for embodied agents with large language models,

    C. H. Song, J. Wu, C. Washington, B. M. Sadler, W.-L. Chao, and Y . Su, “Llm-planner: Few-shot grounded planning for embodied agents with large language models,” in Proceedings of the IEEE/CVF international conference on computer vision , 2023, pp. 2998–3009

    2023

  5. [5]

    Artificial intelligence based object detection and traffic prediction by autonomous vehicles–a review,

    P. Sharma and C. Rana, “Artificial intelligence based object detection and traffic prediction by autonomous vehicles–a review,” Expert Systems with Applications, p. 124664, 2024

    work page 2024

  6. [6]

    Vehicle-road-cloud collaborative perception framework and key technologies: A review,

    B. Gao, J. Liu, H. Zou, J. Chen, L. He, and K. Li, “Vehicle-road-cloud collaborative perception framework and key technologies: A review,” IEEE Transactions on Intelligent Transportation Systems, vol. 25, no. 12, pp. 19 295–19 318, 2024

    work page 2024

  7. [7]

    Embodied intelligence-based perception, decision-making, and control for autonomous operations of rail transportation,

    M. Zhou, H. Dong, H. Song, N. Zheng, W.-H. Chen, and H. Wang, “Embodied intelligence-based perception, decision-making, and control for autonomous operations of rail transportation,” IEEE Transactions on Intelligent Vehicles, pp. 1–5, 2024

    work page 2024

  8. [8]

    Generative diffusion-based contract design for efficient AI twin migration in vehicular embodied AI networks,

    Y . Zhong, J. Kang, J. Wen, D. Ye, J. Nie, D. Niyato, X. Gao, and S. Xie, “Generative diffusion-based contract design for efficient AI twin migration in vehicular embodied AI networks,” IEEE Transactions on Mobile Computing, 2025

    work page 2025

Show all 40 references
  1. [9]

    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

  2. [10]

    Q-IoT: QoS-aware multilayer service architecture for multiclass IoT data traffic management,

    K. C. Serdaroglu, S. Baydere, B. Saovapakhiran, and C. Charnsripinyo, “Q-IoT: QoS-aware multilayer service architecture for multiclass IoT data traffic management,” IEEE Internet of Things Journal , vol. 11, no. 17, pp. 28 330–28 340, 2024

    2024

  3. [11]

    When metaverses meet vehicle road cooperation: Multi-agent DRL-based stackelberg game for vehicular twins migration,

    J. Kang, J. Zhang, H. Yang, D. Ye, and M. S. Hossain, “When metaverses meet vehicle road cooperation: Multi-agent DRL-based stackelberg game for vehicular twins migration,” IEEE Internet of Things Journal , 2024

    2024

  4. [12]

    Learning-based incentive mechanism for task freshness-aware vehic- ular twin migration,

    J. Zhang, J. Nie, J. Wen, J. Kang, M. Xu, X. Luo, and D. Niyato, “Learning-based incentive mechanism for task freshness-aware vehic- ular twin migration,” in 2023 IEEE 43rd International Conference on Distributed Computing Systems Workshops (ICDCSW) . IEEE, 2023, pp. 103–108

    2023

  5. [13]

    Deep reinforcement learning based multi-attribute auction model for resource allocation in vehicular aigc services,

    Y . Tong, J. Kang, J. Chen, M. Xu, H. Yang, M. Xu, R. Xie, and Z. Xiong, “Deep reinforcement learning based multi-attribute auction model for resource allocation in vehicular aigc services,” in Proceedings of the 2024 International Conference on Generative Artificial Intellige...

    2024

  6. [14]

    Blockchain-empowered resource allocation in haps- assisted iov digital twin networks: A federated drl approach,

    H. N. Abishu, A. M. Seid, R. H. Jhaveri, T. R. Gadekallu, A. Erbad, and M. Guizani, “Blockchain-empowered resource allocation in haps- assisted iov digital twin networks: A federated drl approach,” IEEE Transactions on Intelligent Vehicles, 2024

    2024

  7. [15]

    Bidirectional lstm-crf models for sequence tagging,

    Z. Huang, W. Xu, and K. Yu, “Bidirectional lstm-crf models for sequence tagging,” arXiv preprint arXiv:1508.01991 , 2015

    2015 arXiv

  8. [16]

    Finding lottery tickets in vision models via data-driven spectral foresight pruning,

    L. Iurada, M. Ciccone, and T. Tommasi, “Finding lottery tickets in vision models via data-driven spectral foresight pruning,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , 2024, pp. 16 142–16 151

    2024

  9. [17]

    Conversational voice agents are preferred and lead to better driving performance in con- ditionally automated vehicles,

    M. Wang, S. C. Lee, G. Montavon, J. Qin, and M. Jeon, “Conversational voice agents are preferred and lead to better driving performance in con- ditionally automated vehicles,” in Proceedings of the 14th International Conference on Automotive User Interfaces and Interactive Veh...

    2022

  10. [18]

    Aligning cyber space with physical world: A comprehensive survey on embodied AI,

    Y . Liu, W. Chen, Y . Bai, X. Liang, G. Li, W. Gao, and L. Lin, “Aligning cyber space with physical world: A comprehensive survey on embodied AI,” arXiv preprint arXiv:2407.06886 , 2024

    2024 arXiv

  11. [19]

    Embodied understanding of driving scenarios,

    Y . Zhou, L. Huang, Q. Bu, J. Zeng, T. Li, H. Qiu, H. Zhu, M. Guo, Y . Qiao, and H. Li, “Embodied understanding of driving scenarios,” in European Conference on Computer Vision . Springer, 2024, pp. 129– 148

    2024

  12. [20]

    Contextvlm: Zero-shot and few-shot context understanding for autonomous driving using vision language models,

    S. Sural, R. R. Rajkumar et al. , “Contextvlm: Zero-shot and few-shot context understanding for autonomous driving using vision language models,” in 2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC) . IEEE, 2024, pp. 468–475

    2024

  13. [21]

    A compre- hensive survey of few-shot learning: Evolution, applications, challenges, and opportunities,

    Y . Song, T. Wang, P. Cai, S. K. Mondal, and J. P. Sahoo, “A compre- hensive survey of few-shot learning: Evolution, applications, challenges, and opportunities,” ACM Computing Surveys, vol. 55, no. 13s, pp. 1–40, 2023

    2023

  14. [22]

    A fine-grained self-adapting prompt learning approach for few-shot learning with pre-trained language models,

    X. Chen, T. Liu, P. Fournier-Viger, B. Zhang, G. Long, and Q. Zhang, “A fine-grained self-adapting prompt learning approach for few-shot learning with pre-trained language models,” Knowledge-Based Systems, vol. 299, p. 111968, 2024

    2024

  15. [23]

    Primal: Profit maximization avatar placement for mobile edge computing,

    X. Sun and N. Ansari, “Primal: Profit maximization avatar placement for mobile edge computing,” in 2016 IEEE International Conference on Communications (ICC). IEEE, 2016, pp. 1–6

    2016

  16. [24]

    Multiagent deep reinforcement learning for dynamic avatar migration in AIoT-enabled vehicular metaverses with trajectory prediction,

    J. Chen, J. Kang, M. Xu, Z. Xiong, D. Niyato, C. Chen, A. Jamalipour, and S. Xie, “Multiagent deep reinforcement learning for dynamic avatar migration in AIoT-enabled vehicular metaverses with trajectory prediction,” IEEE Internet of Things Journal , vol. 11, no. 1, pp. 70–83, 2024

    2024

  17. [25]

    Stack- elberg game-based multi-agent algorithm for resource allocation and task offloading in mec-enabled c-its,

    S. Zhang, X. Tong, K. Chi, W. Gao, X. Chen, and Z. Shi, “Stack- elberg game-based multi-agent algorithm for resource allocation and task offloading in mec-enabled c-its,” IEEE Transactions on Intelligent Transportation Systems, 2025

    2025

  18. [26]

    Tiny multi-agent DRL for twins migration in UA V metaverses: A multi-leader multi-follower 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 multi-agent DRL for twins migration in UA V metaverses: A multi-leader multi-follower stackelberg game approach,” IEEE Internet of Things Journal , 2024

    2024

  19. [27]

    Pops: Policy pruning and shrinking for deep reinforcement learning,

    D. Livne and K. Cohen, “Pops: Policy pruning and shrinking for deep reinforcement learning,” IEEE Journal of Selected Topics in Signal Processing, vol. 14, no. 4, pp. 789–801, 2020

    2020

  20. [28]

    Qlp: Deep q-learning for pruning deep neural networks,

    E. Camci, M. Gupta, M. Wu, and J. Lin, “Qlp: Deep q-learning for pruning deep neural networks,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 32, no. 10, pp. 6488–6501, 2022

    2022

  21. [29]

    Knowledge distillation based cooperative reinforcement learning for connectivity preservation in UA V networks,

    Z. Wang, Y . Wei, and F. Wu, “Knowledge distillation based cooperative reinforcement learning for connectivity preservation in UA V networks,” in 2021 international conference on UK-China emerging technologies (UCET). IEEE, 2021, pp. 171–176

    2021

  22. [30]

    Five facets of 6G: Research challenges and opportunities,

    L.-H. Shen, K.-T. Feng, and L. Hanzo, “Five facets of 6G: Research challenges and opportunities,” ACM Computing Surveys, vol. 55, no. 11, pp. 1–39, 2023

    2023

  23. [31]

    An overview of otfs for internet of things: Concepts, benefits, and challenges,

    L. Xiao, S. Li, Y . Qian, D. Chen, and T. Jiang, “An overview of otfs for internet of things: Concepts, benefits, and challenges,” IEEE Internet of Things Journal, vol. 9, no. 10, pp. 7596–7618, 2021

    2021

  24. [32]

    Data-aided channel estimation for otfs systems with a superimposed pilot and data transmission scheme,

    W. Yuan, S. Li, Z. Wei, J. Yuan, and D. W. K. Ng, “Data-aided channel estimation for otfs systems with a superimposed pilot and data transmission scheme,” IEEE wireless communications letters , vol. 10, no. 9, pp. 1954–1958, 2021

    1954

  25. [33]

    A mathematical theory of communication,

    C. E. Shannon, “A mathematical theory of communication,” The Bell system technical journal , vol. 27, no. 3, pp. 379–423, 1948

    1948

  26. [34]

    G. T. Fechner, Elemente der psychophysik . Breitkopf u. H ¨artel, 1860, vol. 2

  27. [35]

    Joint user association and resource pricing for metaverse: Distributed and centralized approaches,

    X. Huang, W. Zhong, J. Nie, Q. Hu, Z. Xiong, J. Kang, and T. Q. Quek, “Joint user association and resource pricing for metaverse: Distributed and centralized approaches,” in 2022 IEEE 19th International Confer- ence on Mobile Ad Hoc and Smart Systems (MASS) . IEEE, 2022, pp. 505–513

    2022

  28. [36]

    Stackelberg- game-based computation offloading method in cloud–edge computing networks,

    H. Zhou, Z. Wang, N. Cheng, D. Zeng, and P. Fan, “Stackelberg- game-based computation offloading method in cloud–edge computing networks,” IEEE Internet of Things Journal , vol. 9, no. 17, pp. 16 510– 16 520, 2022

    2022

  29. [37]

    Privacy- preserving incentive mechanism for multi-leader multi-follower IoT- edge computing market: A reinforcement learning approach,

    H. Xu, X. Qiu, W. Zhang, K. Liu, S. Liu, and W. Chen, “Privacy- preserving incentive mechanism for multi-leader multi-follower IoT- edge computing market: A reinforcement learning approach,” Journal of Systems Architecture, vol. 114, p. 101932, 2021

    2021

  30. [38]

    Ntk-guided few-shot class incremental learning,

    J. Liu, Z. Ji, Y . Pang, and Y . Yu, “Ntk-guided few-shot class incremental learning,” IEEE Transactions on Image Processing , 2024

    2024

  31. [39]

    The surprising effectiveness of ppo in cooperative multi-agent games,

    C. Yu, A. Velu, E. Vinitsky, J. Gao, Y . Wang, A. Bayen, and Y . Wu, “The surprising effectiveness of ppo in cooperative multi-agent games,” Advances in neural information processing systems, vol. 35, pp. 24 611– 24 624, 2022

    2022

  32. [40]

    Asynchronous methods for deep rein- forcement learning,

    V . Mnih, A. P. Badia, M. Mirza, A. Graves, T. Lillicrap, T. Harley, D. Silver, and K. Kavukcuoglu, “Asynchronous methods for deep rein- forcement learning,” in International conference on machine learning . PmLR, 2016, pp. 1928–1937

    2016

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.