From federated learning to federated neural architecture search: a survey

Zhu, Hangyu; Zhang, Haoyu; Jin, Yaochu

doi:10.1007/s40747-020-00247-z

Cited by 117 publications

(53 citation statements)

References 92 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Scholars have begun to prioritize automated machine learning (Auto ML), especially neural architecture searching (NAS). Federated NAS requires maximizing the model performance and minimizing the payload to be transferred between the server and clients [32]. Zhu et al [33] regarded it as a multi-objective optimization problem and optimized a neural network model structure for federated learning using a multi-objective evolutionary algorithm.…”

Section: Federated Learningmentioning

confidence: 99%

Transferable face image privacy protection based on federated learning and ensemble models

Yang

Liu

Han

et al. 2021

Complex Intell. Syst.

View full text Add to dashboard Cite

Face image features represent significant user privacy concerns. Face images cannot be privately transferred under existing privacy protection methods, and data across various social networks are unevenly distributed. This paper proposes a method for face image privacy protection based on federated learning and ensemble models. A federated learning model based on distributed data sets was established by means of federated learning. On the client side, a local facial recognition model was obtained by local face data training and used as the input of PcadvGAN to train PcadvGAN for several rounds. On the server side, a parameter aggregator based on a differential evolutionary algorithm was established as the discriminator of PcadvGAN server, and a client facial recognition model was ensembled simultaneously. The discriminator of the PcadvGAN server experienced mutation, crossover, and interaction with the ensemble model to reveal the optimal global weight of the PcadvGAN model. Finally, the global optimal aggregation parameter matrix of PcadvGAN was obtained by calculation. The server and the client shared the global optimal aggregation parameter matrix, enabling each client to generate private face images with high transferability and practicality. Targeted attack and non-targeted attack experiments demonstrated that the proposed method can generate high-quality, transferable, robust, private face images with only minor perturbations more effectively than other existing methods.

show abstract

Section: Federated Learningmentioning

confidence: 99%

Transferable face image privacy protection based on federated learning and ensemble models

Yang

Liu

Han

et al. 2021

Complex Intell. Syst.

View full text Add to dashboard Cite

show abstract

“…Federated learning [33] originates from the attempt to address the privacy concerns in distributed learning [30], which has been widely studied and applied in real-world applications due to its capability of privacy protection and parallel computing [50,60]. Generally, the global optimization objective of a federated learning system can be written by min…”

Section: Federated Learningmentioning

confidence: 99%

“…One exception is our previous work reported in Ref. [51], which proposes a federated data-driven evolutionary algorithm (FDD-EA) for single-objective optimization based on federated learning [33,60]. As its name suggests, federated learning aims to collaboratively train a global model using data distributed on multiple local devices without transmitting the local data to a central server.…”

Section: Introductionmentioning

confidence: 99%

A federated data-driven evolutionary algorithm for expensive multi-/many-objective optimization

Jin

2021

Complex Intell. Syst.

Self Cite

View full text Add to dashboard Cite

Data-driven optimization has found many successful applications in the real world and received increased attention in the field of evolutionary optimization. Most existing algorithms assume that the data used for optimization are always available on a central server for construction of surrogates. This assumption, however, may fail to hold when the data must be collected in a distributed way and are subject to privacy restrictions. This paper aims to propose a federated data-driven evolutionary multi-/many-objective optimization algorithm. To this end, we leverage federated learning for surrogate construction so that multiple clients collaboratively train a radial-basis-function-network as the global surrogate. Then a new federated acquisition function is proposed for the central server to approximate the objective values using the global surrogate and estimate the uncertainty level of the approximated objective values based on the local models. The performance of the proposed algorithm is verified on a series of multi-/many-objective benchmark problems by comparing it with two state-of-the-art surrogate-assisted multi-objective evolutionary algorithms.

show abstract

“…In general, FL approaches have three benefits: (1) data security and privacy is greatly improved without data sharing and centralized data storing [8]; (2) central servers do not need to store massive data, saving memory [9]; and (3) the computation pressure of central servers is reduced as clients in FL approaches undertake part of the model's production training [10]. Additionally, FL approaches are able to supply participants with a higher-quality model, although the volume of each local data set contributing to FL is limited [11].…”

Section: Introductionmentioning

confidence: 99%

Applying Federated Learning in Software-Defined Networks: A Survey

Liao

Zhi

et al. 2022

Symmetry

View full text Add to dashboard Cite

Federated learning (FL) is a type of distributed machine learning approacs that trains global models through the collaboration of participants. It protects data privacy as participants only contribute local models instead of sharing private local data. However, the performance of FL highly relies on the number of participants and their contributions. When applying FL over conventional computer networks, attracting more participants, encouraging participants to contribute more local resources, and enabling efficient and effective collaboration among participants become very challenging. As software-defined networks (SDNs) enable open and flexible networking architecture with separate control and data planes, SDNs provide standardized protocols and specifications to enable fine-grained collaborations among devices. Applying FL approaches over SDNs can take use such advantages to address challenges. A SDN control plane can have multiple controllers organized in layers; the controllers in the lower layer can be placed in the network edge to deal with the asymmetries in the attached switches and hosts, and the controller in the upper layer can supervise the whole network centrally and globally. Applying FL in SDNs with a layered-distributed control plane may be able to protect the data privacy of each participant while improving collaboration among participants to produce higher-quality models over asymmetric networks. Accordingly, this paper aims to make a comprehensive survey on the related mechanisms and solutions that enable FL in SDNs. It highlights three major challenges, an incentive mechanism, privacy and security, and model aggregation, which affect the quality and quantity of participants, the security and privacy in model transferring, and the performance of the global model, respectively. The state of the art in mechanisms and solutions that can be applied to address such challenges in the current literature are categorized based on the challenges they face, followed by suggestions of future research directions. To the best of our knowledge, this work is the first effort in surveying the state of the art in combining FL with SDNs.

show abstract

From federated learning to federated neural architecture search: a survey

Cited by 117 publications

References 92 publications

Transferable face image privacy protection based on federated learning and ensemble models

Transferable face image privacy protection based on federated learning and ensemble models

A federated data-driven evolutionary algorithm for expensive multi-/many-objective optimization

Applying Federated Learning in Software-Defined Networks: A Survey

Contact Info

Product

Resources

About