Federated neural architecture autonomous evolution for collaborative remote sensing scene classification

LI Jianzhao; WANG Shanfeng; YANG Rui; TANG Zedong; ZHOU Yu; GONG Maoguo

doi:10.11834/jrs.20265223

Models and Methods | Views : 0 下载量: 5 CSCD: 0

R-PDF
PDF
Export
Share
Collection
Album

Federated neural architecture autonomous evolution for collaborative remote sensing scene classification
Vol. 30, Issue 5, Pages: 1498-1509(2026)
Received：23 June 2025，

Published：07 May 2026
DOI： 10.11834/jrs.20265223
稿件说明：

移动端阅览

黎健钊，王善峰，杨瑞，唐泽栋，周宇，公茂果.2026.面向神经网络架构自主演进的联邦协同遥感场景分类.遥感学报，30（5）： 1498-1509 DOI： 10.11834/jrs.20265223.

Li J Z，Wang S F，Yang R，Tang Z D，Zhou Y and Gong M G. 2026. Federated neural architecture autonomous evolution for collaborative remote sensing scene classification. National Remote Sensing Bulletin， 30（5）：1498-1509 DOI： 10.11834/jrs.20265223.

摘要

现有联邦遥感场景分类方法通常预设客户端的网络架构固定，忽略了卫星、无人机等设备在算力、存储与能源供应上的显著异构性，导致模型在资源受限终端难以适配部署，限制了联邦系统的实用效率。鉴于此，为探究支持神经网络架构在联邦框架下自主演化的协同方法，本文提出一种基于路径与数据协同采样的神经架构搜索方法。该方法采用“一次训练，多次使用”的神经网络架构搜索训练范式，通过路径与数据双维度协同采样机制实现客户端的个性化适配。在超网的协同训练阶段，各客户端通过梯度范数引导的路径采样动态筛选关键子网架构，基于梯度上界的数据采样聚焦高价值样本，大幅降低超网训练与通信成本；在子网多方部署阶段，通过进化算法基于本地数据搜索验证精度最优的个性化网络架构。此外，在4个基准异构遥感场景分类数据集上的实验验证结果表明，该方法在非独立同分布条件下显著优于固定架构与主流剪枝方法，能在保护数据隐私前提下为异构客户端协同演化出轻量高性能专属网络架构，有效提升了联邦遥感系统的部署可行性与整体性能。

Abstract

Existing federated remote sensing scene classification methods generally assume a fixed client network architecture

disregarding the significant heterogeneity in computing power

storage

and energy supply among devices such as satellites and drones. These limitations make deploying models on resource-constrained terminals difficult

restricting the practical efficiency of federated systems. Therefore

a novel federated remote sensing paradigm that is capable of dynamically adapting to the heterogeneity of client data and computational resources must be urgently developed. While safeguarding the data privacy of all parties

this paradigm should collaboratively explore optimal network architectures that are lighter

more computationally efficient

and tailored to the specific characteristics of each client’s local data. To address this bottleneck

this study proposes a federated collaborative framework with the autonomous evolution of neural network architectures. It proposes a neural architecture search mechanism based on path and data collaborative sampling to achieve lightweight

efficient

and personalized model construction that adapts to local data characteristics. During the collaborative training phase

each client dynamically screens key subnet architectures through gradient norm-guided path sampling and focuses on high-value samples through data sampling based on gradient upper bounds

significantly reducing supernetwork training and communication overhead. In the multiparty deployment phase

evolutionary algorithms are used to search for and validate the personalized subnet with the optimal verification accuracy on the basis of local data. The proposed framework’s foundational breakthrough lies in its integrated approach wherein gradient norm-guided path sampling dynamically identifies and prioritizes architecturally critical subnets during federated training. Simultaneously

gradient-capped data sampling concentrates computational resources on samples with significant effect on training outcomes. These mechanisms collectively form a synergistic strategy that substantially reduces gradient variance and training overhead across the hypernetwork while respecting client resource constraints. Following this collaborative hypernetwork development

each client autonomously executes evolutionary optimization

extracting customized subnetworks that are precisely tailored to local data distributions through adaptation-driven architecture exploration. By unifying adaptive sampling during training with evolutionary personalization while on deployment

the framework achieves unprecedented efficiency in generating lightweight but high-performance models. These models are optimized for diverse edge devices and their unique remote sensing environments while remaining within strict privacy-preserving federated parameters. Experiments on four benchmark heterogeneous remote sensing classification datasets

namely

AID

NWPU-RESISC45

PatternNet

and MEET

demonstrate that this method significantly outperforms fixed architectures and mainstream pruning methods under nonindependent and identically distributed conditions. It can evolve lightweight

high-performance

and dedicated network architectures for heterogeneous clients while protecting data privacy

effectively enhancing the deployment feasibility and overall performance of federated remote sensing systems. Ablation studies have confirmed that the integrated path-data sampling strategy is pivotal to these gains

reducing gradient variance and improving subnet consistency. This work resolves key bottlenecks in federated remote sensing systems by enabling resource-constrained clients to evolve specialized architectures autonomously. The integration of gradient-guided path sampling and data sampling optimizes training efficiency

while evolutionary optimization facilitate personalized subnet deployment. The proposed framework demonstrably enhances deployment feasibility and overall system performance without compromising data privacy

establishing an effective solution for heterogeneous federated learning in remote sensing scene classification.

关键词

Keywords

references

Ben Youssef B , Alhmidi L , Bazi Y and Zuair M . 2024 . Federated learning approach for remote sensing scene classification . Remote Sensing , 16 ( 12 ): 2194 [ DOI: 10.3390/rs16122194 http://dx.doi.org/10.3390/rs16122194 ]

Büyüktaş B , Sumbul G and Demir B . 2024 . Federated learning across decentralized and unshared archives for remote sensing image classification: a review . IEEE Geoscience and Remote Sensing Magazine , 12 ( 3 ): 64 - 80 [ DOI: 10.1109/MGRS.2024.3415391 http://dx.doi.org/10.1109/MGRS.2024.3415391 ]

Chen Z H , Qiu G H , Li P , Zhu L , Yang X K and Sheng B . 2023 . MNGNAS: distilling adaptive combination of multiple searched networks for one-shot neural architecture search . IEEE Transactions on Pattern Analysis and Machine Intelligence , 45 ( 11 ): 13489 - 13508 [ DOI: 10.1109/TPAMI.2023.3293885 http://dx.doi.org/10.1109/TPAMI.2023.3293885 ]

Cheng G , Han J W a nd Lu X Q . 2017 . Remote sensing image scene classification: benchmark and state of the art . Proceedings of the IEEE , 105 ( 10 ): 1865 - 1883 [ DOI: 10.1109/JPROC.2017.2675998 http://dx.doi.org/10.1109/JPROC.2017.2675998 ]

Cheng G , Xie X X , Han J W , Guo L and Xia G S . 2020 . Remote sensing image scene classification meets deep learning: challenges, methods, benchmarks, and opportunities . IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 13 : 3735 - 3756 [ DOI: 10.1109/JSTARS.2020.3005403 http://dx.doi.org/10.1109/JSTARS.2020.3005403 ]

Chu X X , Tian Z , Wang Y Q , Zhang B , Ren H B , Wei X L , Xia H X and Shen C H . 2021 . Twins: revisiting the design of spatial attention in vision transformers // Proceedings of the Annual Conference on Neural Information Processing Systems . Virtual : [s.n.]: 9355 - 9366

He K M , Zhang X Y , Ren S Q and Sun J . 2016 . Deep residual learning for image recognition // Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Las Vegas : IEEE: 770 - 778 [ DOI: 10.1109/CVPR.2016.90 http://dx.doi.org/10.1109/CVPR.2016.90 ]

Hu Z P , Gong M G , Dong Z W , Lu Y H , Li J Z and Zhao Y . 2025 . Heterogeneity-aware pruning framework for personalized federated learning in remote sensing scene classification . Knowledge-Based Systems , 311 : 113007 [ DOI: 10.1016/j.knosys.2025.113007 http://dx.doi.org/10.1016/j.knosys.2025.113007 ]

Li T , Sahu A K , Talwalkar A and Smith V . 2020 . Federated learning: challenges, methods, and future directions . IEEE Signal Processing Magazine , 37 ( 3 ): 50 - 60 [ DOI: 10.1109/MSP.2020.2975749 http://dx.doi.org/10.1109/MSP.2020.2975749 ]

Li J Z , Gong M G , Liu Z T , Wang S F , Zhang Y R , Zhou Y and Gao Y . 2024 . Toward multiparty personalized collaborative learning in remote sensing . IEEE Transactions on Geoscience and Remote Sensing , 62 : 4503616 [ DOI: 10.1109/TGRS.2024.3370584 http://dx.doi.org/10.1109/TGRS.2024.3370584 ]

Li J Z , Wang S F , Yang R , Gong M G , Hu Z P , Zhang N , Sheng K and Zhou Y . 2025a . Toward federated customized neural architecture search for remote sensing scene classification . IEEE Transactions on Geoscience and Remote Sensing , 63 : 4502012 [ DOI: 10.1109/TGRS.2025.3537085 http://dx.doi.org/10.1109/TGRS.2025.3537085 ]

Li Y S , Wu Y N , Cheng G , Tao C , Dang B , Wang Y , Zhang J H , Zhang C G , Liu Y T , Tang X , Ma J Y and Zhang Y J . 2025b . MEET: a million-scale dataset for fine-grained geospatial scene classification with zoom-free remote sensing imagery . IEEE/CAA Journal of Automatica Sinica , 12 ( 5 ): 1004 - 1023 [ DOI: 10.1109/JAS.2025.125324 http://dx.doi.org/10.1109/JAS.2025.125324 ]

Liu J C , Yan J M , Xu H L , Wang Z Y , Huang J Y and Xu Y . 2024 . Finch: enhancing federated learning with hierarchical neural architecture search . IEEE Transactions on Mobile Computing , 23 ( 5 ): 6012 - 6026 [ DOI: 10.1109/TMC.2023.3315451 http://dx.doi.org/10.1109/TMC.2023.3315451 ]

Liu Y Q , Sun Y N , Xue B , Zhang M J , Yen G G and Tan K C . 2023 . A survey on evolutionary neural architecture search . IEEE Transactions on Neural Networks and Learning Systems , 34 ( 2 ): 550 - 570 [ DOI: 10.1109/TNNLS.2021.3100554 http://dx.doi.org/10.1109/TNNLS.2021.3100554 ]

Liu Z , Li J G , Shen Z Q , Huang G , Yan S M and Zhang C S . 2017 . Learning efficient convolutional networks through network slimming // Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV) . Venice : IEEE: 2755 - 2763 [ DOI: 10.1109/ICCV.2017.298 http://dx.doi.org/10.1109/ICCV.2017.298 ]

Lu S , Hu Y , Yang L X , Sun Z H , Mei J L , Tan J C and Song C R . 2023 . PA and DA: jointly sampling path and data for consistent NAS // Proceedings of the 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Vancouver : IEEE: 11940 - 11949 [ DOI: 10.1109/CVPR52729.2023.01149 http://dx.doi.org/10.1109/CVPR52729.2023.01149 ]

Lu Y H , Hu Z P , Zhao W , Guan Z Y , Gong M G and Zhang M Y . 2025 . Layer-interaction adaptive pruning for remote sensing scene classification . IEEE Transactions on Geoscience and Remote Sensing , 63 : 5624712 [ DOI: 10.1109/TGRS.2025.3570653 http://dx.doi.org/10.1109/TGRS.2025.3570653 ]

Lu Z C , Cheng R , Jin Y C , Tan K C and Deb K . 2024 . Neural architecture search as multiobjective optimization benchmarks: problem formulation and performance assessment . IEEE Transactions on Evolutionary Computation , 28 ( 2 ): 323 - 337 [ DOI: 10.1109/TEVC.2022.3233364 http://dx.doi.org/10.1109/TEVC.2022.3233364 ]

Molchanov P , Tyree S , Karras T , Aila T and Kautz J . 2017 . Pruning convolutional neural networks for resource efficient inference . arXiv : 1611 . 06440 [ DOI: 10.48550/arXiv.1611.06440 http://dx.doi.org/10.48550/arXiv.1611.06440 ]

Peng Y M , Song A , Ciesielski V , Fayek H M and Chang X J . 2023 . PRE-NAS: evolutionary neural architecture search with predictor . IEEE Transactions on Evolutionary Computation , 27 ( 1 ): 26 - 36 [ DOI: 10.1109/TEVC.2022.3227562 http://dx.doi.org/10.1109/TEVC.2022.3227562 ]

Qian X L , Li J , Cheng G , Yao X W , Zhao S N , Chen Y B and Jiang L Y . 2018 . Evaluation of the effect of feature extraction strategy on the performance of high-resolution remote sensing image scene classification . Journal of Remote Sensing , 22 ( 5 ): 758 - 776

钱晓亮 , 李佳 , 程塨 , 姚西文 , 赵素娜 , 陈宜滨 , 姜利英 . 2018 . 特征提取策略对高分辨率遥感图像场景分类性能影响的评估 . 遥感学报 , 22 ( 5 ): 758 - 776 [ DOI: 10.11834/jrs.20188015 http://dx.doi.org/10.11834/jrs.20188015 ]

Sandler M , Howard A , Zhu M L , Zhmoginov A and Chen L C . 2018 . MobileNetV2: inverted residuals and linear bottlenecks // Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Salt Lake City : IEEE: 4510 - 4520 [ DOI: 10.1109/CVPR.2018.00474 http://dx.doi.org/10.1109/CVPR.2018.00474 ]

Simonyan K and Zisserman A . 2015 . Very deep convolutional networks for large-scale image recognition . arXiv : 1409 . 1556 [ DOI: 10.48550/arXiv.1409.1556 http://dx.doi.org/10.48550/arXiv.1409.1556 ]

Sun X , Meng Y , Diao W H , Huang L J , Zhang X , Luo J C , Gao L R , Wang P J , Yan Z Y , Gao L J , Dong W , Feng Y C , Li J H and Fu K . 2022 . The review of AI-based intelligent remote sensing capabilities . Journal of Image and Graphics , 27 ( 6 ): 1799 - 1822

孙显 , 孟瑜 , 刁文辉 , 黄丽佳 , 张新 , 骆剑承 , 高连如 , 王佩瑾 , 闫志远 , 郜丽静 , 董文 , 冯瑛超 , 李霁豪 , 付琨 . 2022 . 智能遥感: AI赋能遥感技术 . 中国图象图形学报 , 27 ( 6 ): 1799 - 1822 [ DOI: 10.11834/jig.220161 http://dx.doi.org/10.11834/jig.220161 ]

Xia G S , Hu J W , Hu F , Shi B G , Bai X , Zhong Y F , Zhang L P and Lu X Q . 2017 . AID: a benchmark data set for performance evaluation of aerial scene classification . IEEE Transactions on Geoscience and Remote Sensing , 55 ( 7 ): 3965 - 3981 [ DOI: 10.1109/TGRS.2017.2685945 http://dx.doi.org/10.1109/TGRS.2017.2685945 ]

Yang Q . 2019 . AI and data privacy protection: the way to federated learning . Journal of Information Security Research , 5 ( 11 ): 961 - 965

杨强 . 2019 . AI与数据隐私保护: 联邦学习的破解之道 . 信息安全研究 , 5 ( 11 ): 961 - 965 [ DOI: 10.3969/j.issn.2096-1057.2019.11.003 http://dx.doi.org/10.3969/j.issn.2096-1057.2019.11.003 ]

Yang Z H , Wang Y H , Chen X H , Shi B X , Xu C , Xu C J , Tian Q and Xu C . 2020 . CARS: continuous evolution for efficient neural architecture search // Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Seattle : IEEE: 1826 - 1835 [ DOI: 10.1109/CVPR42600.2020.00190 http://dx.doi.org/10.1109/CVPR42600.2020.00190 ]

Yao Y Q , Cheng G , Xie X X and Han J W . 2021 . Optical remote sensing image object detection based on multi-resolution feature fusion . National Remote Sensing Bulletin , 25 ( 5 ): 1124 - 1137

姚艳清 , 程塨 , 谢星星 , 韩军伟 . 2021 . 多分辨率特征融合的光学遥感图像目标检测 . 遥感学报 , 25 ( 5 ): 1124 - 1137 [ DOI: 10.11834/jrs.20210505 http://dx.doi.org/10.11834/jrs.20210505 ]

Wang S F , Li J Z , Liu Z T , Gong M G , Zhang Y R , Zhao Y , Deng B Y and Zhou Y . 2024 . Personalized multiparty few-shot learning for remote sensing scene classification . IEEE Transactions on Geoscience and Remote Sensing , 62 : 4506115 [ DOI: 10.1109/TGRS.2024.3386978 http://dx.doi.org/10.1109/TGRS.2024.3386978 ]

Zhang C H , Yuan X M , Zhang Q Y , Zhu G X , Cheng L and Zhang N . 2022 . Toward tailored models on private AIoT devices: federated direct neural architecture search . IEEE Internet of Things Journal , 9 ( 18 ): 17309 - 17322 [ DOI: 10.1109/JIOT.2022.3154605 http://dx.doi.org/10.1109/JIOT.2022.3154605 ]

Zhang M , Li H Q , Pan S R , Chang X J , Zhou C , Ge Z Y and Su S . 2021 . One-shot neural architecture search: maximising diversity to overcome catastrophic forgetting . IEEE Transactions on Pattern Analysis and Machine Intelligence , 43 ( 9 ): 2921 - 2935 [ DOI: 10.1109/TPAMI.2020.3035351 http://dx.doi.org/10.1109/TPAMI.2020.3035351 ]

Zhou W X , Newsam S , Li C M and Shao Z F . 2018 . PatternNet: a benchmark dataset for performance evaluation of remote sensing image retrieval . ISPRS Journal of Photogrammetry and Remote Sensing , 145 : 197 - 209 [ DOI: 10.1016/j.isprsjprs.2018.01.004 http://dx.doi.org/10.1016/j.isprsjprs.2018.01.004 ]

Alert me when the article has been cited

提交

Multi-graph convolutional network for a remote sensing image few shot classification

Geomorphological scene classification dataset of high-resolution remote sensing imagery in vegetation-covered areas

CNN-GCN-based dual-stream network for scene classification of remote sensing images