Adaptive vertical federated learning via feature map transferring in mobile edge computing

To bring more intelligence to edge systems, Federated Learning (FL) is proposed to provide a privacy-preserving mechanism to train a globally shared model by utilizing a massive amount of user-generated data on devices. FL enables multiple clients collaboratively train a machine learning model while...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computing Ročník 106; číslo 4; s. 1081 - 1097
Hlavní autoři: Li, Yuanzhang, Sha, Tianchi, Baker, Thar, Yu, Xiao, Shi, Zhiwei, Hu, Sikang
Médium: Journal Article
Jazyk:angličtina
Vydáno: Vienna Springer Vienna 01.04.2024
Springer Nature B.V
Témata:
Algorithms
Artificial Intelligence
Artificial neural networks
Clients
Computer Appl. in Administrative Data Processing
Computer Communication Networks
Computer Science
Datasets
Edge computing
Feature extraction
Feature maps
Federated learning
Image classification
Information Systems Applications (incl.Internet)
Iterative methods
Machine learning
Mobile computing
Representations
Software Engineering
Special Issue Article
Federated learning
68T05
Transfer learning
Convolutional neural network
Machine learning
ISSN:0010-485X, 1436-5057
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:To bring more intelligence to edge systems, Federated Learning (FL) is proposed to provide a privacy-preserving mechanism to train a globally shared model by utilizing a massive amount of user-generated data on devices. FL enables multiple clients collaboratively train a machine learning model while keeping the raw training data local. When the dataset is horizontally partitioned, existing FL algorithms can aggregate CNN models received from decentralized clients. But, it cannot be applied to the scenario where the dataset is vertically partitioned. This manuscript showcases the task of image classification in the vertical FL settings in which participants hold incomplete image pieces of all samples, individually. To this end, the paper discusses AdptVFedConv to tackle this issue and achieves the CNN models’ aim for training without revealing raw data. Unlike conventional FL algorithms for sharing model parameters in every communication iteration, AdptVFedConv enables hidden feature representations. Each client fine-tunes a local feature extractor and transmits the extracted feature representations to the backend machine. A classifier model is trained with concatenated feature representations as input and ground truth labels as output at the server-side. Furthermore, we put forward the model transfer method and replication padding tricks to improve final performance. Extensive experiments demonstrate that the accuracy of AdptVFedConv is close to the centralized model.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0010-485X
1436-5057
DOI:10.1007/s00607-022-01117-x