pith. sign in

arxiv: 2605.25212 · v1 · pith:HGXUFZK2new · submitted 2026-05-24 · 💻 cs.LG · cs.SY· eess.SY

Personalized Federated Learning by Energy-Efficient UAV Communications

Shiqian Guo , Jianqing Liu , Beatriz Lorenzo This is my paper

Pith reviewed 2026-06-30 11:59 UTC · model grok-4.3

classification 💻 cs.LG cs.SYeess.SY
keywords federated learningUAV communicationspersonalized learningenergy efficiencygradient schedulingdevice heterogeneityedge computing
0
0 comments X

The pith

Separating a shared backbone from local heads and selecting only top-gradient devices lets UAVs deliver higher-accuracy federated learning with far lower energy use.

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

The paper establishes a way to run personalized federated learning over UAV links in remote settings where devices hold very different data and UAV batteries are limited. It keeps one backbone model shared across all devices while each device retains its own fixed personalization head. In every round only the devices whose gradients have the largest ℓ2-norm update the backbone. This choice concentrates communication on the most informative contributions and avoids draining the UAV by talking to every device. Simulations indicate the result is both better final accuracy than prior methods and a clear drop in total UAV energy.

Core claim

By enforcing a strict separation between a globally shared backbone and permanently local personalization heads, and by updating the backbone solely with the top-α fraction of devices ranked by gradient ℓ2-norm, the approach reduces the effect of data heterogeneity, focuses optimization on the most useful updates, and produces higher learning accuracy than state-of-the-art schemes while cutting UAV energy consumption.

What carries the argument

The strict separation of a globally shared backbone from device-specific personalization heads, together with gradient ℓ2-norm ranking that restricts backbone updates to the top-α devices.

If this is right

  • Only the most informative device gradients reach the backbone, limiting the slowdown caused by data heterogeneity.
  • UAV energy drops because communication occurs with far fewer devices per round.
  • Final model accuracy exceeds that reported for earlier UAV-assisted or personalized federated learning baselines.
  • The method works without requiring every device to transmit in every round.

Where Pith is reading between the lines

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

  • The same backbone-head split and gradient ranking could be tested on other mobile relays such as ground vehicles or low-orbit satellites.
  • Real-world flight tests that log actual battery drain under the proposed scheduler would show whether the simulated energy savings hold.
  • The selection rule might need adjustment when device participation itself varies over time rather than remaining fixed.

Load-bearing premise

That selecting devices only by the size of their gradient vectors and letting only the largest ones update the shared backbone will keep long-term accuracy high on heterogeneous data while still lowering total energy.

What would settle it

A controlled simulation on highly heterogeneous data in which final test accuracy drops below the accuracy obtained when every device participates, or in which measured UAV energy fails to fall enough to justify any accuracy loss.

Figures

Figures reproduced from arXiv: 2605.25212 by Beatriz Lorenzo, Jianqing Liu, Shiqian Guo.

Figure 1
Figure 1. Figure 1: A UAV-aided FL system with data-heterogeneous [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Robustness of the proposed scheme under device [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 4
Figure 4. Figure 4: Impact of FL on the test accuracy under different [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Comparison of performance under different UAV-aided FL schemes. [PITH_FULL_IMAGE:figures/full_fig_p011_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Comparison of different device scheduling schemes for the case of one local epoch ( [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparison of different device scheduling schemes for the case of three local epochs ( [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Comparison of different schemes with varying [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 10
Figure 10. Figure 10: Performance comparison of selection optimization [PITH_FULL_IMAGE:figures/full_fig_p013_10.png] view at source ↗
read the original abstract

Federated learning (FL) is an effective paradigm for enhancing the learning capability of edge devices while preserving data privacy. In geographically dispersed FL systems, such as sensor networks in remote areas, unmanned aerial vehicles (UAVs) can flexibly establish high-quality communication links to support parameter exchange. However, device heterogeneity and the limited battery capacity of UAVs pose significant challenges. Specifically, data heterogeneity slows convergence, while scheduling all devices for global collaboration incurs excessive communication and energy costs. To overcome these challenges, we adopt a strict separation between a globally shared backbone and permanently local personalization heads, thereby mitigating the impact of data heterogeneity. Furthermore, we propose a gradient-based scheduling strategy that jointly considers energy efficiency and learning performance. In each communication round, the backbone is updated only by the top-$\alpha$ devices ranked by gradient $\ell_{2}$-norm, ensuring that optimization focuses on the most informative updates. Simulation results demonstrate that the proposed scheme achieves higher learning accuracy than state-of-the-art approaches while significantly reducing UAV energy consumption.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a personalized federated learning framework for UAV-assisted systems that separates a shared backbone from local personalization heads to handle data heterogeneity, and introduces a gradient ℓ₂-norm based scheduling policy that selects only the top-α fraction of devices for backbone updates in each round. Simulation results are claimed to show superior learning accuracy compared to state-of-the-art methods alongside reduced UAV energy consumption.

Significance. If the simulation-based claims hold under rigorous controls for heterogeneity and long-term convergence, the work could offer a practical approach to balancing personalization, convergence speed, and energy efficiency in UAV-enabled FL deployments. The explicit separation of backbone and heads is a clear strength for mitigating non-IID effects, and the energy-aware scheduling is a relevant contribution for resource-constrained aerial platforms.

major comments (2)
  1. [Simulation Results] Simulation Results section: the central performance claims rest entirely on unspecified simulations; no details are provided on baselines, data heterogeneity levels (e.g., Dirichlet parameter or class imbalance), UAV channel models, number of independent runs, or statistical significance tests, so the evidence for higher accuracy and lower energy cannot be evaluated.
  2. [Gradient-based scheduling strategy] Gradient-based scheduling strategy (described in the abstract and method): ranking devices solely by ||∇||₂ and updating the backbone with only the top-α fraction lacks any analysis or ablation showing that this rule avoids selection bias on non-IID partitions; high-norm gradients may simply reflect local loss curvature or outlier statistics rather than representative information, risking drift of the shared backbone away from the full data distribution over long horizons.
minor comments (1)
  1. [Abstract] Abstract: the claim that the scheduling strategy 'jointly considers energy efficiency and learning performance' is not supported by the stated rule (ranking performed solely by gradient ℓ₂-norm); clarify whether energy enters the ranking criterion or is handled separately.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments. We address each major point below and will revise the manuscript to strengthen the presentation of results and analysis.

read point-by-point responses

  1. Referee: [Simulation Results] Simulation Results section: the central performance claims rest entirely on unspecified simulations; no details are provided on baselines, data heterogeneity levels (e.g., Dirichlet parameter or class imbalance), UAV channel models, number of independent runs, or statistical significance tests, so the evidence for higher accuracy and lower energy cannot be evaluated.

    Authors: We agree that the Simulation Results section requires substantially more detail for the claims to be properly evaluated. In the revised manuscript we will expand this section to specify: all baselines and their hyper-parameters, the exact data heterogeneity generation process (Dirichlet concentration parameter together with class-imbalance ratios), the UAV channel model (path-loss exponent, small-scale fading distribution, and noise variance), the number of independent Monte-Carlo runs (minimum of five) with reported means and standard deviations, and the statistical tests (paired t-tests or Wilcoxon signed-rank tests) used to support accuracy and energy-consumption differences. revision: yes

  2. Referee: [Gradient-based scheduling strategy] Gradient-based scheduling strategy (described in the abstract and method): ranking devices solely by ||∇||₂ and updating the backbone with only the top-α fraction lacks any analysis or ablation showing that this rule avoids selection bias on non-IID partitions; high-norm gradients may simply reflect local loss curvature or outlier statistics rather than representative information, risking drift of the shared backbone away from the full data distribution over long horizons.

    Authors: The manuscript states that the scheduler jointly accounts for energy efficiency and learning performance, yet the referee correctly notes the absence of any ablation or bias analysis. We will add (i) an ablation comparing gradient-norm selection against random and loss-value selection across multiple Dirichlet heterogeneity levels, (ii) long-horizon convergence plots that track backbone drift relative to the full population distribution, and (iii) a short discussion of the conditions under which high-norm gradients may be unrepresentative. These additions will be placed in a new subsection of the method and in the experimental evaluation. revision: yes

Circularity Check

0 steps flagged

No circularity: method is a proposed design choice validated by simulation, with no equations or self-referential reductions.

full rationale

The paper presents a separation of backbone and personalization heads plus a top-α gradient-norm scheduling rule as explicit design choices, not as quantities derived from prior fitted parameters or self-citations that reduce the claimed accuracy/energy gains back to the inputs by construction. No equations, uniqueness theorems, or ansatzes are invoked in the provided text; the central claims rest on empirical simulation comparisons rather than any mathematical chain that collapses to tautology. This is the normal case of an engineering proposal whose validity is external to its own definitions.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review supplies no explicit free parameters, axioms, or invented entities; α (the scheduling fraction) is mentioned but its value and fitting procedure are not stated.

pith-pipeline@v0.9.1-grok · 5708 in / 1046 out tokens · 19821 ms · 2026-06-30T11:59:35.155742+00:00 · methodology

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

41 extracted references · 2 canonical work pages · 1 internal anchor

  1. [1]

    Learning in the air: Secure federated learning for uav- assisted crowdsensing

    Yuntao Wang, Zhou Su, Ning Zhang, and Abderrahim Bensli- mane. Learning in the air: Secure federated learning for uav- assisted crowdsensing. IEEE Transactions on Network Science and Engineering, 8(2):1055–1069, 2021. IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. X, NO. X, X X 15

    2021

  2. [2]

    Privacy-preserving federated learning for uav-enabled networks: Learning-based joint scheduling and resource management

    Helin Yang, Jun Zhao, Zehui Xiong, Kwok-Yan Lam, Sumei Sun, and Liang Xiao. Privacy-preserving federated learning for uav-enabled networks: Learning-based joint scheduling and resource management. IEEE Journal on Selected Areas in Communications, 39(10):3144–3159, 2021

    2021

  3. [3]

    Energy-efficient federated learning over uav- enabled wireless powered communications

    Quoc-Viet Pham, Mai Le, Thien Huynh-The, Zhu Han, and Won-Joo Hwang. Energy-efficient federated learning over uav- enabled wireless powered communications. IEEE Transactions on Vehicular Technology, 71(5):4977–4990, 2022

    2022

  4. [4]

    Empowering edge intelligence by air-ground integrated federated learning

    Yuben Qu, Chao Dong, Jianchao Zheng, Haipeng Dai, Fan Wu, Song Guo, and Alagan Anpalagan. Empowering edge intelligence by air-ground integrated federated learning. IEEE Network, 35(5):34–41, 2021

    2021

  5. [5]

    Heterogeneous personalized federated learning by local-global updates mixing via convergence rate

    Meirui Jiang, Anjie Le, Xiaoxiao Li, and Qi Dou. Heterogeneous personalized federated learning by local-global updates mixing via convergence rate. In The Twelfth International Conference on Learning Representations, 2024

    2024

  6. [6]

    Federated Learning with Personalization Layers

    Manoj Ghuhan Arivazhagan, Vinay Aggarwal, Aaditya Kumar Singh, and Sunav Choudhary. Federated learning with person- alization layers. arXiv preprint arXiv:1912.00818, 2019

    work page internal anchor Pith review Pith/arXiv arXiv 1912

  7. [7]

    Over-the-air federated learning in digital twins empowered uav swarms

    Bingqing Jiang, Jun Du, Chunxiao Jiang, Zhu Han, Ahmed Alhammadi, and Merouane Debbah. Over-the-air federated learning in digital twins empowered uav swarms. IEEE Trans- actions on Wireless Communications, 2024

    2024

  8. [8]

    Energy and gradient aware dynamic scheduling for v2v aided federated edge learning

    Shiqian Guo and Bin-Jie Hu. Energy and gradient aware dynamic scheduling for v2v aided federated edge learning. IEEE Communications Letters, 28(2):323–327, 2023

    2023

  9. [9]

    Energy-efficient federated learning through uav edge under location uncertainties

    Chen Wang, Xiao Tang, Daosen Zhai, Ruonan Zhang, Nurzhan Ussipov, and Yan Zhang. Energy-efficient federated learning through uav edge under location uncertainties. IEEE Transac- tions on Network Science and Engineering, 2024

    2024

  10. [10]

    Effective uav-aided asyn- chronous decentralized federated learning with distributed, adaptive and energy-aware gradient sparsification

    Alia Asheralieva and Dusit Niyato. Effective uav-aided asyn- chronous decentralized federated learning with distributed, adaptive and energy-aware gradient sparsification. IEEE In- ternet of Things Journal, 2025

    2025

  11. [11]

    Density-clustering aggregation for personalized federated learning with ai-enabled aerial and edge computing in uavs

    Wei-Che Chien, Chih-Hsun Lin, Tianli Zhu, Cheng Dai, Sahil Garg, and Amrit Mukherjee. Density-clustering aggregation for personalized federated learning with ai-enabled aerial and edge computing in uavs. IEEE Internet of Things Journal, 2025

    2025

  12. [12]

    Decentralized navigation with heterogeneous federated rein- forcement learning for uav-enabled mobile edge computing

    Pengfei Wang, Hao Yang, Guangjie Han, Ruiyun Yu, Leyou Yang, Geng Sun, Heng Qi, Xiaopeng Wei, and Qiang Zhang. Decentralized navigation with heterogeneous federated rein- forcement learning for uav-enabled mobile edge computing. IEEE Transactions on Mobile Computing, 2024

    2024

  13. [13]

    Integrated sensing, computation, and communication for uav-assisted federated edge learning

    Yao Tang, Guangxu Zhu, Wei Xu, Man Hon Cheung, Tat-Ming Lok, and Shuguang Cui. Integrated sensing, computation, and communication for uav-assisted federated edge learning. IEEE Transactions on Wireless Communications, 2025

    2025

  14. [14]

    Multi-uav hierarchical intelligent traffic offloading network optimization based on deep federated learning

    Fengqi Li, Kaiyang Zhang, Jiaheng Wang, Yudong Li, Fengqiang Xu, Yanjuan Wang, and Ning Tong. Multi-uav hierarchical intelligent traffic offloading network optimization based on deep federated learning. IEEE Internet of Things Journal, 11(12):21312–21324, 2024

    2024

  15. [15]

    Age of information minimization in uav-enabled iot net- works via federated reinforcement learning

    Xuhui Zhang, Huijun Xing, Yanyan Shen, Jie Xu, and Shuguang Cui. Age of information minimization in uav-enabled iot net- works via federated reinforcement learning. IEEE Transactions on Wireless Communications, 2025

    2025

  16. [16]

    Trajectory control and fair communications for multi-uav networks: A federated multi-agent deep reinforcement learning approach

    Getaneh Berie Tarekegn, Belayneh Abebe Tesfaw, Rong-Terng Juang, Dola Saha, Robel Berie Tarekegn, Hsin-Piao Lin, and Li-Chia Tai. Trajectory control and fair communications for multi-uav networks: A federated multi-agent deep reinforcement learning approach. IEEE Transactions on Wireless Communi- cations, 2025

    2025

  17. [17]

    Uav-assisted unbiased hierarchical federated learning: Performance and convergence analysis

    Ruslan Zhagypar, Nour Kouzayha, Hesham ElSawy, Hayssam Dahrouj, and Tareq Y Al-Naffouri. Uav-assisted unbiased hierarchical federated learning: Performance and convergence analysis. IEEE Transactions on Machine Learning in Commu- nications and Networking, 2025

    2025

  18. [18]

    The data value based asynchronous federated learning for uav swarm under unstable communication scenarios

    Zhenhua Cui, Tao Yang, Xiaofeng Wu, Hui Feng, and Bo Hu. The data value based asynchronous federated learning for uav swarm under unstable communication scenarios. IEEE Transactions on Mobile Computing, 23(6):7165–7179, 2024

    2024

  19. [19]

    Three-stage stackelberg game enabled clustered feder- ated learning in heterogeneous uav swarms

    Wenji He, Haipeng Yao, Tianle Mai, Fu Wang, and Mohsen Guizani. Three-stage stackelberg game enabled clustered feder- ated learning in heterogeneous uav swarms. IEEE Transactions on Vehicular Technology, 72(7):9366–9380, 2023

    2023

  20. [20]

    Multi-uav-assisted federated learning for energy-aware distributed edge training

    Jianhang Tang, Jiangtian Nie, Yang Zhang, Zehui Xiong, Wen- chao Jiang, and Mohsen Guizani. Multi-uav-assisted federated learning for energy-aware distributed edge training. IEEE Transactions on Network and Service Management, 21(1):280– 294, 2024

    2024

  21. [21]

    Zoheb Hassan, and Md

    Mohammed Saif, Md. Zoheb Hassan, and Md. Jahangir Hossain. Decentralized aggregation for energy-efficient federated learn- ing in mmwave aerial-terrestrial integrated networks. IEEE Transactions on Machine Learning in Communications and Networking, 2:1283–1304, 2024

    2024

  22. [22]

    Communication-efficient learning of deep networks from decentralized data

    Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hamp- son, and Blaise Aguera y Arcas. Communication-efficient learning of deep networks from decentralized data. In Artificial intelligence and statistics, pages 1273–1282. PMLR, 2017

    2017

  23. [23]

    Towards personalized federated learning

    Alysa Ziying Tan, Han Yu, Lizhen Cui, and Qiang Yang. Towards personalized federated learning. IEEE Transactions on Neural Networks and Learning Systems, 34(12):9587–9603, 2023

    2023

  24. [24]

    Personalized federated learning with model-contrastive learning for multi- modal user modeling in human-centric metaverse

    Xiaokang Zhou, Qiuyue Yang, Xuzhe Zheng, Wei Liang, Kevin I-Kai Wang, Jianhua Ma, Yi Pan, and Qun Jin. Personalized federated learning with model-contrastive learning for multi- modal user modeling in human-centric metaverse. IEEE Journal on Selected Areas in Communications, 42(4):817–831, 2024

    2024

  25. [25]

    Shuai Wang, Yanqing Xu, Yanli Yuan, and Tony Q. S. Quek. Toward fast personalized semi-supervised federated learning in edge networks: Algorithm design and theoretical guarantee. IEEE Transactions on Wireless Communications, 23(2):1170– 1183, 2024

    2024

  26. [26]

    Decentralized navigation with heterogeneous federated rein- forcement learning for uav-enabled mobile edge computing

    Pengfei Wang, Hao Yang, Guangjie Han, Ruiyun Yu, Leyou Yang, Geng Sun, Heng Qi, Xiaopeng Wei, and Qiang Zhang. Decentralized navigation with heterogeneous federated rein- forcement learning for uav-enabled mobile edge computing. IEEE Transactions on Mobile Computing, 23(12):13621–13638, 2024

    2024

  27. [27]

    Joint uav deployment and resource allocation: A personalized federated deep reinforcement learning approach

    Xinyi Xu, Gang Feng, Shuang Qin, Yijing Liu, and Yao Sun. Joint uav deployment and resource allocation: A personalized federated deep reinforcement learning approach. IEEE Trans- actions on Vehicular Technology, 73(3):4005–4018, 2024

    2024

  28. [28]

    Quantization-aware collaborative inference for large embodied ai models

    Zhonghao Lyu, Ming Xiao, Mikael Skoglund, Merouane Debbah, and H Vincent Poor. Quantization-aware collaborative inference for large embodied ai models. arXiv preprint arXiv:2602.13052, 2026

    work page arXiv 2026

  29. [29]

    Energy-efficient uav-assisted federated learning: Trajec- tory optimization, device scheduling, and resource management

    Zhenyu Fu, Juan Liu, Yuyi Mao, Long Qu, Lingfu Xie, and Xijun Wang. Energy-efficient uav-assisted federated learning: Trajec- tory optimization, device scheduling, and resource management. IEEE Transactions on Network and Service Management, 2025

    2025

  30. [30]

    Joint uav deployment and edge association for energy-efficient federated learning

    Tao Wu, Maomao Li, Yuben Qu, Hongjun Wang, Zhenhua Wei, and Jiannong Cao. Joint uav deployment and edge association for energy-efficient federated learning. IEEE Transactions on Cognitive Communications and Networking, 2025

    2025

  31. [31]

    Energy efficient and low latency federated distillation over uav-assisted wireless networks

    Zhe Zhang, Yanchao Zhao, Chuyi Chen, Kun Zhu, and Dusit Niyato. Energy efficient and low latency federated distillation over uav-assisted wireless networks. IEEE Transactions on Wireless Communications, 2025

    2025

  32. [32]

    Energy- efficient distributed learning for noma-based unmanned aerial agent-assisted mec networks

    Prakhar Consul, Ishan Budhiraja, Deepak Garg, Neeraj Kumar, Joel JPC Rodrigues, and Abdullah Almuhaideb. Energy- efficient distributed learning for noma-based unmanned aerial agent-assisted mec networks. IEEE Internet of Things Journal, 2025

    2025

  33. [33]

    Distributed optimization of resource- efficiency for federated edge intelligence in uav-enabled iot networks

    Yingyu Li, Jiangying Rao, Lian Wang, Yong Xiao, Xiaohu Ge, and Guangming Shi. Distributed optimization of resource- efficiency for federated edge intelligence in uav-enabled iot networks. IEEE Internet of Things Journal, 2025

    2025

  34. [34]

    Joint communication and computing resource allocation for energy efficient hierarchical federated learning in marine internet of things

    Wei Jiang, Zhongjie Xiao, Li Ping Qian, Shuang Qin, Gang Feng, and Yuan Wu. Joint communication and computing resource allocation for energy efficient hierarchical federated learning in marine internet of things. IEEE Transactions on Network Science and Engineering, 2025

    2025

  35. [35]

    Energy minimization for wireless communication with rotary-wing uav

    Yong Zeng, Jie Xu, and Rui Zhang. Energy minimization for wireless communication with rotary-wing uav. IEEE Transac- tions on Wireless Communications, 18(4):2329–2345, 2019

    2019

  36. [36]

    Cellular uav-to-x communications: Design and optimization for multi-uav networks

    Shuhang Zhang, Hongliang Zhang, Boya Di, and Lingyang Song. Cellular uav-to-x communications: Design and optimization for multi-uav networks. IEEE Transactions on Wireless Communi- cations, 18(2):1346–1359, 2019

    2019

  37. [37]

    Mobile unmanned aerial vehicles (uavs) IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL

    Mohammad Mozaffari, Walid Saad, Mehdi Bennis, and Mérouane Debbah. Mobile unmanned aerial vehicles (uavs) IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. X, NO. X, X X 16 for energy-efficient internet of things communications. IEEE Transactions on Wireless Communications, 16(11):7574–7589, 2017

    2017

  38. [38]

    Learning multiple layers of features from tiny images

    Alex Krizhevsky, Geoffrey Hinton, et al. Learning multiple layers of features from tiny images. Toronto, ON, Canada, 2009

    2009

  39. [39]

    Reliable and interpretable personalized federated learning

    Zixuan Qin, Liu Yang, Qilong Wang, Yahong Han, and Qinghua Hu. Reliable and interpretable personalized federated learning. In 2023 IEEE/CVF Conference on Computer Vision and Pat- tern Recognition (CVPR), pages 20422–20431, 2023

    2023

  40. [40]

    Intelligent uav deployment for a disaster-resilient wireless network

    Hassaan Hydher, Dushantha Nalin K Jayakody, Kasun T Hemachandra, and Tharaka Samarasinghe. Intelligent uav deployment for a disaster-resilient wireless network. Sensors, 20(21):6140, 2020

    2020

  41. [41]

    Linear convergence of gradient and proximal-gradient methods under the polyak-łojasiewicz condition

    Hamed Karimi, Julie Nutini, and Mark Schmidt. Linear convergence of gradient and proximal-gradient methods under the polyak-łojasiewicz condition. In Joint European conference on machine learning and knowledge discovery in databases, pages 795–811. Springer, 2016. Shiqian Guo received the Ph.D. degree in information and communication engineering from Sout...

    2016