Federated Learning via Over-the-Air Computation
Open accessThe DOI matches its Crossref record; 61 of the 62 references it cites were found in a registry or an index.
Abstract
The stringent requirements for low-latency and privacy of the emerging high-stake applications with intelligent devices such as drones and smart vehicles make the cloud computing inapplicable in these scenarios. Instead, edge machine learning becomes increasingly attractive for performing training and inference directly at network edges without sending data to a centralized data center. This stimulates a nascent field termed as federated learning for training a machine learning model on computation, storage, energy and bandwidth limited mobile devices in a distributed manner. To preserve data privacy and address the issues of unbalanced and non-IID data points across different devices, the federated averaging algorithm has been proposed for global model aggregation by computing the weighted average of locally updated model at each selected device. However, the limited communication bandwidth becomes the main bottleneck for aggregating the locally computed updates. We thus propose a novel over-the-air computation based approach for fast global model aggregation via exploring the superposition property of a wireless multiple-access channel. This is achieved by joint device selection and beamforming design, which is modeled as a sparse and low-rank optimization problem to support efficient algorithms design. To achieve this goal, we provide a difference-of-convex-functions (DC) representation for the sparse and low-rank function to enhance sparsity and accurately detect the fixe…
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“For instance, over-the-air federated learning (FL) provides a collaborative ML framework to train a global statistical model over wireless networks without accessing edge devices' private raw data [31].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“Specifically, the transmitted signals at edge devices are first multiplied by the fading channels and then superposed over-the-air with additive channel noise, resulting in a noisy weighted sum of transmitted signals [31].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“Specifically, for edge training systems, we need to jointly optimize the subcarrier and bandwidth allocation [88], [90], [241], transmit power and receive beamforming [31], [48], [178], passive beamforming at RIS [48], [178], device selection [31], [242] and activity detection [76], local updates computation [92], and global aggregation frequency control [130], thereby reducing the optimality gap”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“For instance, the group sparsity can represent the combinatorial variables for edge devices selection in FL [31], edge devices activity detection [76], and inference tasks selection [53].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
In the citing paper
“Most of the resource allocation problems in edge AI need to solve a series of nonconvex optimization problems, e.g., nonconvex sparse optimization for device selection in wireless FL, nonconvex quadratic programming for transceiver design in over-the-air FL [31], lowrank matrix optimization for interference management in edge device distributed inference [52], and unit modulus constrained phase”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [31]
2 more claims about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Hao, X.-B., & Zhang, G.-G. (2017) · Encycl. Semantic Comput. Robotic Intell. · doi:10.1142/s2425038416300184 · Cited by 86,401 · cited 1×
“THE astounding growth in data volume promotes widespread artificial intelligent applications such as image recognition and natural language processing [1], thanks to”
BackgroundLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Boyd, S., & Vandenberghe, L. (2004) · doi:10.1017/cbo9780511804441 · Cited by 56,601
Vorontsova, E., Hildebrand, R., Gasnikov, A. V., & Stonyakin, F. (2010) · Computer Vision · doi:10.1007/978-1-4614-5257-7_6 · Cited by 39,270 · cited 2×
“The resulting problem is known as the sum-of-infeasibilities in the literature of optimization [39].”
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
Communication-Efficient Learning of Deep Networks from Decentralized Data
McMahan, H. B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. Y. (2016) · arXiv · doi:10.48550/arxiv.1602.05629 · Cited by 14,215 · cited 7×
“In particular, the federated averaging (FedAvg) algorithm [12] turns out to be a promising way to efficiently average the locally updated model at each device with unbalanced and non-IID data, thereby reducing the number of communication rounds between the center node and the end devices.”
InfluentialFederated Machine Learning: Concept and Applications
Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3298981 · Cited by 6,322
Han, S., Mao, H., & Dally, W. J. (2015) · arXiv · doi:10.48550/arxiv.1510.00149 · Cited by 4,662 · cited 1×
“To address this issue, there is a growing body of recent works to reduce the storage overhead, time and power consumption in the inference process using the model compression methods via hardware and software co-design [7], [8].”
MethodTowards Federated Learning at Scale: System Design
Bonawitz, K., Eichner, H., Grieskamp, W., Huba, D., Ingerman, A., Ivanov, V., Kiddon, C., Konečný, J., Mazzocchi, S., McMahan, H. B., et al. (2019) · arXiv · doi:10.48550/arxiv.1902.01046 · Cited by 2,334 · cited 2×
“It has promising applications [14], [15] such as smart retail, smart healthcare, financial services, mobile content predictions, etc.”
BackgroundFederated Learning with Non-IID Data
Zhao, Y., Li, M., Lai, L., Suda, N., Civin, D., & Chandra, V. (2018) · arXiv · doi:10.48550/arxiv.1806.00582 · Cited by 2,077 · cited 1×
“, the data is generated by distinct distributions across different devices), imposes significant statistical challenges to fit a mode from the non-IID data [13], [17].”
Background
Research works printed but not linked: 2
- rockafellar, 2015, Convex Analysis
- tao, 1997, Convex analysis approach to DC programming: Theory, algorithms and applications, Acta Math Vietn, 22, 289
Cited by
Federated Learning in Mobile Edge Networks: A Comprehensive Survey
Lim, W. Y. B., Luong, N. C., Hoang, D. T., Jiao, Y., Liang, Y.-C., Yang, Q., Niyato, D., & Miao, C. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2986024 · Cited by 2,565
“Similar to [129], the authors in [130] propose an integration of computation and communication via over-the-air computation.”
BackgroundA survey on federated learning
Zhang, C., Xie, Y., Bai, H., Yu, B., Li, W., & Gao, Y. (2021) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.106775 · Cited by 1,899
Federated Learning for Healthcare Informatics
Xu, J., Glicksberg, B. S., Su, C., Walker, P., Bian, J., & Wang, F. (2019) · arXiv · doi:10.48550/arxiv.1911.06270 · Cited by 1,538
“Problems Paper Indices Statistical [110] [132] [89] [79] [129] [37] [27] [34] [52] [23] [45] Communication [64] [72] [92] [107] [87] [72] [16] [24] [3] [99] [15] [133] [48] [5] [105] [117] [126]”
BackgroundConvergence of Edge Computing and Deep Learning: A Comprehensive Survey
Wang, X., Han, Y., Leung, V. C. M., Niyato, D., Yan, X., & Chen, X. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2970550 · Cited by 1,498
“However, this synchronous updating mode potentially cannot scale well, and is inefficient and inflexible in view of two key properties of FL, specifically, 1) infrequent training tasks, since edge devices typically have weaker computing power and limited battery endurance and thus cannot afford intensive training task…”
BackgroundInfluentialA Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection
Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., Liu, X., & He, B. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3124599 · Cited by 1,407
Energy Efficient Federated Learning Over Wireless Communication Networks
Yang, Z., Chen, M., Saad, W., Hong, C. S., & Shikh-Bahaei, M. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3037554 · Cited by 1,153
“To improve the statistical learning performance for on-device distributed training, the authors in [33] developed a novel sparse and low-rank modeling approach.”
MethodEdge Intelligence: The Confluence of Edge Computing and Artificial Intelligence
Deng, S., Zhao, H., Fang, W., Yin, J., Dustdar, S., & Zomaya, A. Y. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2984887 · Cited by 1,147
“ion property of multi-access channels [38]. The proposed BAA can dramatically reduce communication latency compared with traditional Orthogonal Frequency Division Multiple Access (OFDMA). The work in [39] explores the over-the-air computation for model aggregation in Federated Learning. More specifically, Yang et al. p…”
MethodEdge artificial intelligence for 6g: vision, enabling technologies, and applications
Letaief, K. B., Shi, Y., Lu, J., & Lu, J. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3126076 · Cited by 829
“Section II-B: Wireless Techniques for Edge Training Over-the-Air Computation Low-latency analog model aggregation [31], [44], [45]”
BackgroundMethodInfluentialSelf-citationAbdulrahman, S., Tout, H., Ould-Slimane, H., Mourad, A., Talhi, C., & Guizani, M. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3030072 · Cited by 796
“As the bottleneck of federated averaging method lies in the restricted communication band-width that delays the clients from uploading their updates, the work in [44] proposes faster aggregation model.”
MethodA Survey on Federated Learning for Resource-Constrained IoT Devices
Imteaj, A., Thakker, U., Wang, S., Li, J., & Amini, M. H. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3095077 · Cited by 774
Xu, W., Yang, Z., Ng, D. W. K., Levorato, M., Eldar, Y. C., & Debbah, M. (2023) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2023.3239189 · Cited by 660
“Different from these methods, the works [146], [164] developed a unified difference-of-convex-functions (DC) programming based”
InfluentialScheduling Policies for Federated Learning in Wireless Networks
Yang, H. H., Liu, Z., Quek, T. Q. S., & Poor, H. V. (2020) · IEEE Transactions on Communications · doi:10.1109/tcomm.2019.2944169 · Cited by 640
“When spectral resources become the communication bottleneck, there are new methods exploiting the computeover-air mechanism and a jointly decode-and-average scheme at the edge computing unit [17], [18].”
BackgroundMethodFederated Learning Over Wireless Fading Channels
Amiri, M. M., & Gunduz, D. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.2974748 · Cited by 625
“Then we studied the alternative analog transmission approach, cosidered recently in [25]–[27], which does not employ quantization or channel coding, and exploits the superposition property of the wireless MAC, rather than orthogonalizing the transmissions from different devices.”
MethodMachine Learning at the Wireless Edge: Distributed Stochastic Gradient Descent Over-the-Air
Amiri, M. M., & Gunduz, D. (2019) · 2019 IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2019.8849334 · Cited by 109
“While our work can be easily extended to the fading model, the distinctive contributions of our work with respect to the concurrent works of [40] and [41] are i) the introduction of a purely digital separate gradient compression and communication scheme; ii) the consideration of a bandwidth-constrained channel, which…”
BackgroundDistributed Learning in Wireless Networks: Recent Progress and Future Challenges
Chen, M., Gunduz, D., Huang, K., Saad, W., Bennis, M., Feljan, A. V., & Poor, H. V. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118346 · Cited by 548
“As we discussed previously, researchers have also designed AirComp FEEL systems over multiple-antenna channels [38], [112], [113].”
MethodMachine Learning at the Wireless Edge: Distributed Stochastic Gradient Descent Over-the-Air
Mohammadi Amiri, M., & Gunduz, D. (2020) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2020.2981904 · Cited by 548
“We also remark that both [32] and [33] consider transmitting model updates, while we focus on gradient transmission, which is more energyefficient as each device transmits only the innovation obtained through gradient descent at that particular iteration (together with error accumulation), whereas model transmission w…”
BackgroundWu, Q., Xu, J., Zeng, Y., Ng, D. W. K., Al-Dhahir, N., Schober, R., & Swindlehurst, A. L. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3088681 · Cited by 562
Latency Minimization for Intelligent Reflecting Surface Aided Mobile Edge Computing
Bai, T., Pan, C., Deng, Y., Elkashlan, M., Nallanathan, A., & Hanzo, L. (2020) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2020.3007035 · Cited by 528
“[18] conceived an over-the-air computation aided federated learning algorithm for reducing both the latency and the power consumption, and for preserving the users’ privacy in MEC systems.”
BackgroundCommunication-efficient edge ai: algorithms and systems
Shi, Y., Yang, K., Jiang, T., Zhang, J., & Letaief, K. B. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.3007787 · Cited by 488
“The connection between data aggregation from distributed nodes in edge training and the in-network computation problem [20] in wireless sensor networks has been established in [21], which proposed an over-the-air computation approach for fast model”
BackgroundMethodInfluentialSelf-citationXu, J., & Wang, H. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3031503 · Cited by 466
“Recognizing the unique physical property of wireless transmission, [2], [20]–[22] propose analog model aggregation over the air, provided that a very stringent synchronization is available.”
BackgroundInfluentialWahab, O. A., Mourad, A., Otrok, H., & Taleb, T. (2021) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2021.3058573 · Cited by 487
“The authors of [101] propose an over-the-air approach for efficient global parameters aggregation.”
BackgroundHierarchical Federated Learning ACROSS Heterogeneous Cellular Networks
Abad, M. S. H., Ozfatura, E., GUndUz, D., & Ercetin, O. (2020) · ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp40776.2020.9054634 · Cited by 424
Federated Learning for Vehicular Internet of Things: Recent Advances and Open Issues
Du, Z., Wu, C., Yoshinaga, T., Yau, K.-L. A., Ji, Y., & Li, J. (2020) · IEEE Open Journal of the Computer Society · doi:10.1109/ojcs.2020.2992630 · Cited by 393
“[57] propose an over-the-air computation-based aggregation approach for FL in wireless environments.”
BackgroundJoint Device Scheduling and Resource Allocation for Latency Constrained Wireless Federated Learning
Shi, W., Zhou, S., Niu, Z., Jiang, M., & Geng, L. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3025446 · Cited by 381
“ting FL in wireless networks has received many research efforts. To reduce the uploading latency introduced by global model aggregation, novel analog aggregation techniques have been proposed in [15]–[17]. For analog aggregation, the scheduled devices concurrently transmit their local models via analog modulation in a…”
MethodZhu, G., Du, Y., Gunduz, D., & Huang, K. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3039309 · Cited by 376
“Existing works on over-the-air aggregation, as well as its application in the context of FEEL [6], [7], [17]–[19] consider analog modulation assuming that the transmitter can modulate the carrier waveform as desired, freely choosing the I/Q coefficients as arbitrary real number.”
BackgroundMethodInfluentialFederated Learning for 6G: Applications, Challenges, and Opportunities
Yang, Z., Chen, M., Wong, K.-K., Poor, H. V., & Cui, S. (2022) · Engineering · doi:10.1016/j.eng.2021.12.002 · Cited by 374
“To overcome this challenge, an emerging approach is over-the-air computation, which can gather wireless data quickly by using the superposition nature of wireless transmission [34,35].”
Collaborative Federated Learning for Healthcare: Multi-Modal COVID-19 Diagnosis at the Edge
Qayyum, A., Ahmad, K., Ahsan, M. A., Al-Fuqaha, A., & Qadir, J. (2022) · IEEE Open Journal of the Computer Society · doi:10.1109/ojcs.2022.3206407 · Cited by 345
“The edge computing network is potentially more heterogeneous as compared to any other central network and clients in an edge computing network vary due to data acquisition resources [56], such as communication, computational, and storage resources, etc.”
BackgroundA Survey of Beam Management for mmWave and THz Communications Towards 6G
Xue, Q., Ji, C., Ma, S., Guo, J., Xu, Y., Chen, Q., & Zhang, W. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3361991 · Cited by 280
“As an alternative, by exploiting the waveform superposition property, over-the-air computation (AirComp) has recently been integrated into FL to quickly aggregate local models so as to improve the communication efficiency [101], [105], [132].”
MethodHolistic Network Virtualization and Pervasive Network Intelligence for 6G
Shen, X., Gao, J., Wu, W., Li, M., Zhou, C., & Zhuang, W. (2022) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2021.3135829 · Cited by 392
“Over-the-air computation based approaches are investigated in [194], [216], [217].”
MethodStochastic Controlled Averaging for Federated Learning with Communication Compression
Huang, X., Li, P., & Li, X. (2023) · International Conference on Learning Representations · doi:10.48550/arxiv.2308.08165 · Cited by 35
Dynamic-Fusion-Based Federated Learning for COVID-19 Detection
Zhang, W., Zhou, T., Lu, Q., Wang, X., Zhu, C., Sun, H., Wang, Z., Lo, S. K., & Wang, F.-Y. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2021.3056185 · Cited by 329
Digital Twin of Wireless Systems: Overview, Taxonomy, Challenges, and Opportunities
Khan, L. U., Han, Z., Saad, W., Hossain, E., Guizani, M., & Hong, C. S. (2022) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2022.3198273 · Cited by 322
“privacy and homomorphic encryption, few works [120], [121], [122] proposed over-the-air computation.”
BackgroundMulti-Armed Bandit-Based Client Scheduling for Federated Learning
Xia, W., Quek, T. Q. S., Guo, K., Wen, W., Yang, H. H., & Zhu, H. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3008091 · Cited by 313
“Recognizing the superposition property of wireless multiple-access channels, computation-over-the-air approaches which harness interference instead of suppressing interference are shown to be able to accelerate global model aggregation, such as broadband analog aggregation adopted in [7]–[9].”
BackgroundLim, W. Y. B., Huang, J., Xiong, Z., Kang, J., Niyato, D., Hua, X.-S., Leung, C., & Miao, C. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3056341 · Cited by 318
Optimizing Federated Learning in Distributed Industrial IoT: A Multi-Agent Approach
Zhang, W., Yang, D., Wu, W., Peng, H., Zhang, N., Zhang, H., & Shen, X. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118352 · Cited by 300
“[20] aimed at maximizing the number of participating devices in each FL epoch via exploiting superposition property of wireless multiple-access channels.”
BackgroundA Survey on Integrated Sensing, Communication, and Computation
Wen, D., Zhou, Y., Li, X., Shi, Y., Huang, K., & Letaief, K. B. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3521498 · Cited by 363
“Although transmitting model parameters reduce the communication overhead by a significant amount compared to transmitting raw data, scarce communication resources still limit the communication efficiency in wireless FL systems, which motivates the emergence of AirComp-based FL [22], [23], [78], [79].”
Self-citationLim, W. Y. B., Ng, J. S., Xiong, Z., Jin, J., Zhang, Y., Niyato, D., Leung, C., & Miao, C. (2022) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2021.3096076 · Cited by 312
“Several works have proposed a variety of solutions, e.g., model compression techniques such as quantization and subsampling [17], client selection protocols to reduce the occurrences of stragglers [13], as well as Broadband Analog Aggregation (BAA) with over-the-air computation [18].”
BackgroundSplit Learning Over Wireless Networks: Parallel Design and Resource Management
Wu, W., Li, M., Qu, K., Zhou, C., Shen, X., Zhuang, W., Li, X., & Shi, W. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3242704 · Cited by 278
“Extensive works are devoted to optimizing FL performance from different research directions, such as multi-tier FL framework design to accommodate a large number of devices [6], [7], and model aggregation and compression techniques to reduce communication overhead [8], [9].”
BackgroundWhat is Semantic Communication? A View on Conveying Meaning in the Era of Machine Intelligence
Lan, Q., Wen, D., Zhang, Z., Zeng, Q., Chen, X., Popovski, P., & Huang, K. (2021) · Journal of Communications and Information Networks · doi:10.23919/jcin.2021.9663101 · Cited by 255
“• Distributed learning : Local gradient computation [64-68], over-the-air computing [69-72], importance-aware RRM [73-76], differential privacy [77,78].”
BackgroundEnergy-Efficient Radio Resource Allocation for Federated Edge Learning
Zeng, Q., Du, Y., Huang, K., & Leung, K. K. (2020) · 2020 IEEE International Conference on Communications Workshops (ICC Workshops) · doi:10.1109/iccworkshops49005.2020.9145118 · Cited by 286
“One branch of research focuses on edge-device selection so as to accelerate learning [5], [6].”
BackgroundAdaptive Federated Learning and Digital Twin for Industrial Internet of Things
Sun, W., Lei, S., Wang, L., Liu, Z., & Zhang, Y. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.3034674 · Cited by 327
“[7] proposed a fast global aggregation scheme by device selection and beamforming design.”
BackgroundMethodInfluentialUVeQFed: Universal Vector Quantization for Federated Learning
Shlezinger, N., Chen, M., Eldar, Y. C., Poor, H. V., & Cui, S. (2021) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2020.3046971 · Cited by 275
“This property of wireless communications is known to affect the design of deep learning systems operating over such channels [41]–[47].”
BackgroundEdge Intelligence: Empowering Intelligence to the Edge of Network
Xu, D., Li, T., Li, Y., Su, X., Tarkoma, S., Jiang, T., Crowcroft, J., & Hui, P. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2021.3119950 · Cited by 272
Self-citationZhu, G., Lyu, Z., Jiao, X., Liu, P., Chen, M., Xu, J., Cui, S., & Zhang, P. (2023) · Science China Information Sciences · doi:10.1007/s11432-022-3652-2 · Cited by 239
“The work [65] presented a joint design of device scheduling and receive beamforming in which the edge devices with weak signal strengths after receive beamforming were dropped from the training process.”
MethodOver-the-Air Federated Learning From Heterogeneous Data
Sery, T., Shlezinger, N., Cohen, K., & Eldar, Y. (2021) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2021.3090323 · Cited by 219
“Unlike previous convergence proofs of OTA FL, our analysis of COTAF is not restricted to sparsified updates as in [20] or to full gradient descent optimization as in [21], and holds for the typical FL setting with SGD-based training and heterogeneous data.”
MethodOver-the-Air Computing for Wireless Data Aggregation in Massive IoT
Zhu, G., Xu, J., Huang, K., & Cui, S. (2021) · IEEE Wireless Communications · doi:10.1109/mwc.011.2000467 · Cited by 235
“To solve the problem intelligently, AirComp has been recently developed as a new air-interface solution for fast model update aggregation in distributed edge learning [10], [12], [13].”
MethodOptimized Power Control for Over-the-Air Computation in Fading Channels
Cao, X., Zhu, G., Xu, J., & Huang, K. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3012287 · Cited by 248
“Recently, the applications of AirComp have expanded from sensor networks towards networks supporting edge machine learning [4], [5].”
BackgroundInfluentialPrivacy for Free: Wireless Federated Learning via Uncoded Transmission With Adaptive Power Control
Liu, D., & Simeone, O. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2020.3036948 · Cited by 234
Scheduling in Cellular Federated Edge Learning with Importance and Channel AwarenessPreprint
Ren, J., He, Y.-H., Wen, D.-Z., Yu, G., Huang, K.-B., & Guo, D. (2020) · arXiv.org · doi:10.48550/arxiv.2004.00490 · Cited by 228
“Recent years have seen much research interests from both industry and academia, covering key topics such as learning update compression [16]–[18], radio resource management [19]–[23], and over-the-air computation [24]–[26].”
BackgroundConvergence of Update Aware Device Scheduling for Federated Learning at the Wireless Edge
Amiri, M. M., Gunduz, D., Kulkarni, S. R., & Poor, H. V. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3052681 · Cited by 232
“the number of participating devices and overcoming the lack of CSI at the devices are introduced in [13] and [14], respectively.”
BackgroundTask-Oriented Communications for 6G: Vision, Principles, and Technologies
Shi, Y., Zhou, Y., Wen, D., Wu, Y., Jiang, C., & Letaief, K. B. (2023) · IEEE Wireless Communications · doi:10.1109/mwc.002.2200468 · Cited by 189
Self-citationStacked Intelligent Metasurface-Aided MIMO Transceiver Design
An, J., Yuen, C., Xu, C., Li, H., Ng, D. W. K., Di Renzo, M., Debbah, M., & Hanzo, L. (2024) · IEEE Wireless Communications · doi:10.1109/mwc.013.2300259 · Cited by 173
“To aggregate the distributed trained models, over-the-air computation (AirComp) approaches have shown promise by leveraging the signal superposition property of wireless multiple-access channels [15].”
MethodZeng, Q., Du, Y., Huang, K., & Leung, K. K. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3088910 · Cited by 214
“ervices. Designing techniques for suppressing the overhead forms a vein of active research, called communication-efficient FEEL. Diversified approaches have been proposed such as device scheduling [6], [7], designing customized multi-access technologies [2], [3], and optimizing uploading frequencies [4]. The second chal…”
BackgroundMachine Learning for Large-Scale Optimization in 6G Wireless Networks
Shi, Y., Lian, L., Shi, Y., Wang, Z., Zhou, Y., Fu, L., Bai, L., Zhang, J., & Zhang, W. (2023) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2023.3300664 · Cited by 205
“FL [157], [158] has been proposed as a prominent distributed ML scheme to effectively solve the model optimization problem in ML over large-scale wireless networks, where each edge device participates in the training process by exchanging model parameters with data kept locally.”
MethodSelf-citationA Survey on Over-the-Air Computation
Şahin, A., & Yang, R. (2023) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2023.3264649 · Cited by 162
“The utilized precoder and combining vectors are similar to the ones with multiple antennas [113] since the effective channel can be expressed as a diagonal MIMO matrix.”
BackgroundMethodWireless Federated Learning with Local Differential Privacy
Seif, M., Tandon, R., & Li, M. (2020) · 2020 IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit44484.2020.9174426 · Cited by 200
Federated Learning via Intelligent Reflecting Surface
Wang, Z., Qiu, J., Zhou, Y., Shi, Y., Fu, L., Chen, W., & Letaief, K. B. (2022) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3099505 · Cited by 205
“Therefore, we have ∂Mh1 = ∂Mh3 = ρ · ∂‖M‖2 and ∂V h2 = ∂V h4 = ∂‖V ‖2, where ∂‖X‖2 can be efficiently computed by u1u H 1 [17] and u1 is the eigenvector corresponding to the largest eigenvalue of the matrix X .”
BackgroundMethodSelf-citationAn Incentive Mechanism for Federated Learning in Wireless Cellular Networks: An Auction Approach
Thi Le, T. H., Tran, N. H., Tun, Y. K., Nguyen, M. N. H., Pandey, S. R., Han, Z., & Hong, C. S. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3062708 · Cited by 198
“the fairness issue. The study in [18] recommended combining deep reinforcement learning (DRL) and FL frameworks with mobile edge systems to optimize computing, caching and communication. The study in [19] jointly considered the device selection and beamforming fast global model aggregation. They used the principle of…”
BackgroundApplications of federated learning in smart cities: recent advances, taxonomy, and open challenges
Zheng, Z., Zhou, Y., Sun, Y., Wang, Z., Liu, B., & Li, K. (2022) · Connection Science · doi:10.1080/09540091.2021.1936455 · Cited by 206
“For example, to reduce time complexity, the FedAvg algorithm (K. Yang et al., 2020) is used for local calculation updates and aggregation; it is also used in client-side differential privacy preservation federated optimisation algorithms.”
MethodLi, L., Shi, D., Hou, R., Li, H., Pan, M., & Han, Z. (2021) · IEEE INFOCOM 2021 - IEEE Conference on Computer Communications · doi:10.1109/infocom42981.2021.9488839 · Cited by 196
“in [5] developed a fast model aggregation approach with joint device selection and beamforming design, which considers only one communication round, and thus cannot guarantee the longterm training performance.”
BackgroundNomikos, N., Gkonis, P. K., Bithas, P. S., & Trakadas, P. (2023) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2022.3225590 · Cited by 201
“In cases where data privacy issues arise, resorting to federated learning is preferable as the federated averaging algorithm builds a global model by aggregating the weighted average of locally updated model at each network device [151].”
MethodOptimized Power Control Design for Over-the-Air Federated Edge Learning
Cao, X., Zhu, G., Xu, J., Wang, Z., & Cui, S. (2022) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3126060 · Cited by 173
“Prior works studied the Air-FEEL system from different perspectives such as devices scheduling [22]– [25], beamforming design [26], [27], update compression [28]–[30], and hyper-parameters (such as learning rates) optimization [31]–[33].”
InfluentialCell-Free Massive MIMO for Wireless Federated Learning
Vu, T. T., Ngo, D. T., Tran, N. H., Ngo, H. Q., Dao, M. N., & Middleton, R. H. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.3002988 · Cited by 194
“It is worth noting that the existing works [10], [16] rely on an impractical assumption that the channel state information (CSI) remains unchanged during the whole FL process.”
BackgroundBruce Lee-Inspired Fluid Antenna System: Six Research Topics and the Potentials for 6G
Wong, K., Tong, K., Shen, Y., Chen, Y., & Zhang, Y. (2022) · Frontiers in Communications and Networks · doi:10.3389/frcmn.2022.853416 · Cited by 201
Age-Based Scheduling Policy for Federated Learning in Mobile Edge Networks
Yang, H. H., Arafa, A., Quek, T. Q. S., & Vincent Poor, H. (2020) · ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp40776.2020.9053740 · Cited by 189
“…studies have been carried out and resulted in various scheduling protocols for FL, ranging from minimizing the transmission latency [13], maximizing the spectral utility [14], to opportunistically alternating between selecting the UEs with advantageous and disadvantageous channel conditions [15].”
BackgroundMethodWireless Communications for Collaborative Federated Learning
Chen, M., Poor, H. V., Saad, W., & Cui, S. (2020) · IEEE Communications Magazine · doi:10.1109/mcom.001.2000397 · Cited by 192
“Therefore, it is necessary to consider the optimization of wireless networks to improve FL performance, as pointed out in [7]–[9].”
BackgroundFederated Machine Learning for Intelligent IoT via Reconfigurable Intelligent Surface
Yang, K., Shi, Y., Zhou, Y., Yang, Z., Fu, L., & Chen, W. (2020) · IEEE Network · doi:10.1109/mnet.011.2000045 · Cited by 184
“Based on these two key observations of federated machine learning, a joint device selection and receiver beamforming design problem is formulated in [6] for fast model aggregation of federated machine learning, thereby simultaneously improving the model accuracy and model training speed.”
BackgroundMethodSelf-citationHierarchical Incentive Mechanism Design for Federated Machine Learning in Mobile Networks
Lim, W. Y. B., Xiong, Z., Miao, C., Niyato, D., Yang, Q., Leung, C., & Poor, H. V. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2985694 · Cited by 233
“With FL, only the model parameters need to be updated by the model owners [11].”
BackgroundFederated Learning: A signal processing perspective
Gafni, T., Shlezinger, N., Cohen, K., Eldar, Y. C., & Poor, H. V. (2022) · IEEE Signal Processing Magazine · doi:10.1109/msp.2021.3125282 · Cited by 176
6G-enabled Edge AI for Metaverse: Challenges, Methods, and Future Research Directions
Chang, L., Zhang, Z., Li, P., Xi, S., Guo, W.-X., Shen, Y.-K., Xiong, Z.-H., Kang, J., Niyato, D., Qiao, X.-Q., et al. (2022) · J. Commun. Inf. Networks · doi:10.48550/arxiv.2204.06192 · Cited by 176
Hu, S., Chen, X., Ni, W., Hossain, E., & Wang, X. (2021) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2021.3086014 · Cited by 178
“The data are normalized to be a zero-mean Gaussian process with a unit variance [151], [152].”
BackgroundMethodOn Analog Gradient Descent Learning Over Multiple Access Fading Channels
Sery, T., & Cohen, K. (2020) · IEEE Transactions on Signal Processing · doi:10.1109/tsp.2020.2989580 · Cited by 172
“In [15], [16], multiple antennas were used at the receiver, where beamforming was used to maximize the number of devices scheduled for transmission.”
MethodCommunication and computation efficiency in Federated Learning: A survey
Almanifi, O. R. A., Chow, C.-O., Tham, M.-L., Chuah, J. H., & Kanesan, J. (2023) · Internet of Things · doi:10.1016/j.iot.2023.100742 · Cited by 161
Lo, S. K., Lu, Q., Wang, C., Paik, H.-Y., & Zhu, L. (2022) · ACM Computing Surveys · doi:10.1145/3450288 · Cited by 202
“ral server: Physically, the central servers are usually hosted on a local machine [89, 194], cloud server [57, 155, 178], mobile edge computing platform [122], edge gateway [147, 207] or base station [9, 66, 107, 184]. Most studies propose the central server as a hardware component that creates the first global model…”
BackgroundMethodInfluentialLiu, H., Yuan, X., & Zhang, Y.-J. A. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3086116 · Cited by 165
“Moreover, the authors in [8] maximized the number of participants in over-the-air model aggregation subject to a specific communication error constraint.”
BackgroundMethodInfluentialFederated Dropout—A Simple Approach for Enabling Federated Learning on Resource Constrained Devices
Wen, D., Jeon, K. J., & Huang, K. (2022) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2022.3149783 · Cited by 178
“To improve the communication efficiency, researchers have proposed new communication techniques including over-the-air aggregation [4]–[6], joint computation-and-communication resource management [7]– [10], and high-dimensional source encoding [11].”
MethodFederated Learning in Smart Cities: A Comprehensive SurveyPreprint
Zheng, Z., Zhou, Y., Sun, Y.-L., Wang, Z., Liu, B., & Li, K.-Q. (2021) · arXiv.org · doi:10.48550/arxiv.2102.01375 · Cited by 168
“For example, reduce time complexity[116], FedAvg algorithm[117] is used for local calculation update and aggregation, and used for client-side differen-tial privacy preservation federal optimization algorithm[35].”
MethodGradient Statistics Aware Power Control for Over-the-Air Federated Learning
Zhang, N., & Tao, M. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3065748 · Cited by 153
“Therein, a channel-inversion like power control scheme, similar to those in [18], [19], [22] is designed so that the gradient vectors sent from the selected devices are aligned at the edge server.”
BackgroundMethodInfluentialLearning Rate Optimization for Federated Learning Exploiting Over-the-Air Computation
Xu, C., Liu, S., Yang, Z., Huang, Y., & Wong, K.-K. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118402 · Cited by 159
“The aggregate error is further reduced by additionally considering DLR, compared to the methods utilizing only on wireless resource in [26], [35], which validates Theorem 2 and Theorem 3.”
BackgroundMethodEnergy-Efficient Federated Edge Learning with Joint Communication and Computation Design
Mo, X., & Xu, J. (2021) · Journal of Communications and Information Networks · doi:10.23919/jcin.2021.9475121 · Cited by 164
Integrated Sensing, Communication, and Computation Over-the-Air: MIMO Beamforming Design
Li, X., Liu, F., Zhou, Z., Zhu, G., Wang, S., Huang, K., & Gong, Y. (2023) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2022.3233795 · Cited by 168
“The existing literature have investigated the implementation of AirComp in FEEL from different perspectives, including communication-learning tradeoff [16], devices scheduling [17], update compression [18], beamforming design [19], hyper-parameters control [20], learning rate control [21], power control [22], digital…”
BackgroundMethodCyber Threat Intelligence Sharing Scheme Based on Federated Learning for Network Intrusion Detection
Sarhan, M., Layeghy, S., Moustafa, N., & Portmann, M. (2023) · Journal of Network and Systems Management · doi:10.1007/s10922-022-09691-3 · Cited by 175
Intelligent Reflecting Surface for Downlink Non-Orthogonal Multiple Access Networks
Fu, M., Zhou, Y., & Shi, Y. (2019) · 2019 IEEE Globecom Workshops (GC Wkshps) · doi:10.1109/gcwkshps45667.2019.9024675 · Cited by 199
“Although the SDR technique can solve problems (11) and (15), the probability of the obtained solution being rank-one is small, especially when the dimension of the optimization parameters is high [9, 18].”
BackgroundSelf-citationSemi-Decentralized Federated Learning With Cooperative D2D Local Model Aggregations
Lin, F. P.-C., Hosseinalipour, S., Azam, S. S., Brinton, C. G., & Michelusi, N. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3118344 · Cited by 140
Hu, X., Masouros, C., & Wong, K. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2021.3066495 · Cited by 158
“Note that one subgradient of ||Ψ||s at point Ψ can be efficiently computed as z1z1 , where z1 is the vector corresponding to the largest singular value of Ψ [35].”
Dynamic Aggregation for Heterogeneous Quantization in Federated Learning
Chen, S., Shen, C., Zhang, L., & Tang, Y. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3076613 · Cited by 132
“Joint learning, wireless resource allocation, and user selection schemes for FL are studied in [7], [8], [28]–[30].”
BackgroundGuo, H., Liu, A., & Lau, V. K. N. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3002925 · Cited by 105
“The tradeoff between the number of active clients and the SNR is investigated in [14] and [15].”
BackgroundChallenges, Applications and Design Aspects of Federated Learning: A Survey
Rahman, K. M. J., Ahmed, F., Akhter, N., Hasan, M., Amin, R., Aziz, K. E., Islam, A. K. M. M., Mukta, M. S. H., & Islam, A. K. M. N. (2021) · IEEE Access · doi:10.1109/access.2021.3111118 · Cited by 147
“ment by using superposition property of multiple-access channel [38]–[40]; asynchronous communication [41] for”
MethodFederated Learning and Wireless Communications
Qin, Z., Li, G. Y., & Ye, H. (2021) · IEEE Wireless Communications · doi:10.1109/mwc.011.2000501 · Cited by 177
“The effective approach in [9] selects a subset of local clients with proper transmission power and carefully adjusts the beamforming vector to optimize the statistical performance of federated learning.”
BackgroundMethodInfluentialDu, J., Jiang, B., Jiang, C., Shi, Y., & Han, Z. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3242727 · Cited by 167
“Instead of maximizing the average number of selected devices per round as studied in [41], [42], Lemma 2 provides us with a reasonable direction to minimize the upper bound of E [f (wt)]−E [f (wt−1)] alternatively to resist the channel fading and noise.”
BackgroundSelf-citationCommunication-efficient distributed learning: An overview
Cao, X., Başar, T., Diggavi, S., Eldar, Y. C., Letaief, K. B., Poor, H. V., & Zhang, J. (2023) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2023.3242710 · Cited by 139
“[157] to reduce the bandwidth requirement by exploiting the superposition property of wireless multiple-access channels.”
BackgroundQuantized Federated Learning Under Transmission Delay and Outage Constraints
Wang, Y., Xu, Y., Shi, Q., & Chang, T. (2021) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2021.3126081 · Cited by 156
Reconfigurable Intelligent Surface Empowered Downlink Non-Orthogonal Multiple Access
Fu, M., Zhou, Y., Shi, Y., & Letaief, K. B. (2021) · IEEE Transactions on Communications · doi:10.1109/tcomm.2021.3066587 · Cited by 175
“over-the-air computation for federated learning [47].”
Self-citationLevel-5 Autonomous Driving—Are We There Yet? A Review of Research Literature
Khan, M. A., Sayed, H. E., Malik, S., Zia, T., Khan, J., Alkaabi, N., & Ignatious, H. A. (2022) · ACM Computing Surveys · doi:10.1145/3485767 · Cited by 172
Semisupervised Federated-Learning-Based Intrusion Detection Method for Internet of Things
Zhao, R., Wang, Y., Xue, Z., Ohtsuki, T. O., Adebisi, B., & Gui, G. (2022) · IEEE Internet of Things Journal · doi:10.1109/jiot.2022.3175918 · Cited by 177
Wireless-Powered Over-the-Air Computation in Intelligent Reflecting Surface-Aided IoT Networks
Wang, Z., Shi, Y., Zhou, Y., Zhou, H., & Zhang, N. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3015489 · Cited by 131
“As the principle of AirComp matches well with the model aggregation process in federated learning, the authors in [11]–[14] leveraged the advantages of AirComp to accelerate the convergence rate of federated learning over wireless networks.”
BackgroundSelf-citationTopology-aware Federated Learning in Edge Computing: A Comprehensive Survey
Wu, J., Dong, F., Leung, H., Zhu, Z., Zhou, J., & Drew, S. (2024) · ACM Computing Surveys · doi:10.1145/3659205 · Cited by 130
“In each round of FL, updates were validated and packaged by the selected committee, allowing the most honest nodes to improve the global model continuously.”
BackgroundMethodFederated Edge Learning with Misaligned Over-The-Air Computation
Shao, Y., Gündüz, D., & Liew, S. C. (2021) · IEEE International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) · doi:10.1109/spawc51858.2021.9593155 · Cited by 19
“Prior works on OAC considered only the perfectly aligned case [10], [11], [13], [14], [19] where there is neither channelgain misalignment nor time misalignment, which we refer to as the aligned OAC.”
BackgroundEfficient Federated Learning Algorithm for Resource Allocation in Wireless IoT Networks
Nguyen, V.-D., Sharma, S. K., Vu, T. X., Chatzinotas, S., & Ottersten, B. (2021) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.3022534 · Cited by 136
“The authors in [25] studied a sparse and low-rank problem to support FL algorithms, where user selection and beamforming design are jointly optimized.”
BackgroundAl-Quraan, M. M., Mohjazi, L., Bariah, L., Centeno, A., Zoha, A., Arshad, K., Assaleh, K. T., Muhaidat, S., Debbah, M. A., & Imran, M. A. (2023) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2023.3251404 · Cited by 186
“To this end, AirComp has been proposed with the aim to provide a co-design approach for the FL aggregation procedure, by utilising superposition nature of radio channels for simultaneously transmitting multiple models updates from different clients [182].”
Method
Metadata
- Authors
- Yang, K., Jiang, T., Shi, Y., & Ding, Z.
- Affiliations
- Chinese Academy of Sciences (CN)
- ShanghaiTech University (CN)
- University of California, Davis (US)
- Venue
- IEEE Transactions on Wireless Communications
- Published
- 2020-03
- Volume
- 19
- Issue
- 3
- Pages
- 2022-2035
- ISSN
- 1536-1276 (ISSN-L)
- 1558-2248
- OpenAlex
- W2999074226
- Cited by
- 1,084
- References
- 62
- Citations by year
- 2026: 86 · 2025: 154 · 2024: 176 · 2023: 190 · 2022: 189 · 2021: 190 · 2020: 80
- FWCI
- 63.79
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Wireless Communications (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (61601290)
- National Science Foundation (CNS-1702752, ECCS-1711823)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Wireless Communication Technologies
- Keywords
- federated learning, over-the-air computation, edge machine learning, model aggregation, device selection, beamforming design, difference of convex functions, DC algorithm, non-IID data, data privacy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.