29,333 Hits in 4.3 sec

Federated Learning of a Mixture of Global and Local Models [article]

Filip Hanzely, Peter Richtárik
2021 arXiv   pre-print
We propose a new optimization formulation for training federated learning models.  ...  Notably, our methods are similar but not identical to federated averaging / local SGD, thus shedding some light on the role of local steps in federated learning.  ...  Instead of learning a single global model by solving (1), we propose to learn a mixture of the global model and the purely local models which can be trained by each device i using its data D i only.  ... 
arXiv:2002.05516v3 fatcat:mijaykh7rje55b5x2nr2cwxdxu

Specialized federated learning using a mixture of experts [article]

Edvin Listo Zec, Olof Mogren, John Martinsson, Leon René Sütfeld, Daniel Gillblad
2021 arXiv   pre-print
To achieve this personalization we propose a federated learning framework using a mixture of experts to combine the specialist nature of a locally trained model with the generalist knowledge of a global  ...  In federated learning, clients share a global model that has been trained on decentralized local client data.  ...  Figure 1 . 1 Federated mixtures of experts, consisting of a global model fg and local specialist models f k s using local gating functions h k .  ... 
arXiv:2010.02056v3 fatcat:6bku362pl5fjjf62mhgp5hajou

Subspace Learning for Personalized Federated Optimization [article]

Seok-Ju Hahn, Minwoo Jeong, Junghye Lee
2021 arXiv   pre-print
i.e. global model and local model).  ...  Training federated learning systems usually focuses on optimizing a global model that is identically deployed to all client devices.  ...  between a local and a global model.  ... 
arXiv:2109.07628v1 fatcat:aogce7pjavd33od6ka4ktc4h2y

Certified Robustness in Federated Learning [article]

Motasem Alfarra, Juan C. Pérez, Egor Shulgin, Peter Richtárik, Bernard Ghanem
2022 arXiv   pre-print
Finally, we explore the robustness of mixtures of global and local~(\ie personalized) models, and find that the robustness of local models degrades as they diverge from the global model  ...  We show several advantages of personalization over both~(that is, only training on local data and federated training) in building more robust models with faster training.  ...  At last, we analyze a version of the recently-proposed mixture between global and personalized models from federated learning [19, 20] from a certified robustness lens.  ... 
arXiv:2206.02535v1 fatcat:3tshzwpowfcypifr3rvvq2wklm

Survey of Personalization Techniques for Federated Learning [article]

Viraj Kulkarni, Milind Kulkarni, Aniruddha Pant
2020 arXiv   pre-print
The standard formulation of federated learning produces one shared model for all clients.  ...  Federated learning enables machine learning models to learn from private decentralized data without compromising privacy.  ...  Instead of learning a single global model, each device learns a mixture of the global model and its own local model.  ... 
arXiv:2003.08673v1 fatcat:6uqguovedvfghh4v6dwkvtyt6q

Separate but Together: Unsupervised Federated Learning for Speech Enhancement from Non-IID Data [article]

Efthymios Tzinis, Jonah Casebeer, Zhepei Wang, Paris Smaragdis
2021 arXiv   pre-print
Each client trains their model in isolation using mixture invariant training while periodically providing updates to a central server.  ...  We simulate a real-world scenario where each client only has access to a few noisy recordings from a limited and disjoint number of speakers (hence non-IID).  ...  We analyze the convergence of the global model learned in a federated setup and we compare it against training on each one of the private datasets of the C individual nodes.  ... 
arXiv:2105.04727v3 fatcat:o5p4ig6fljaotacm66c5gmjdb4

Federated Learning: A Signal Processing Perspective [article]

Tomer Gafni, Nir Shlezinger, Kobi Cohen, Yonina C. Eldar, H. Vincent Poor
2021 arXiv   pre-print
Federated learning is an emerging machine learning paradigm for training models across multiple edge devices holding local datasets, without explicitly exchanging the data.  ...  We present a formulation for the federated learning paradigm from a signal processing perspective, and survey a set of candidate approaches for tackling its unique challenges.  ...  set of global models, i.e., a mixture of models (also referred to as model interpolation [61] ).  ... 
arXiv:2103.17150v2 fatcat:pktgiqowsjbklfnj753ehdbnhu

Collaborative Prognostics for Machine Fleets Using a Novel Federated Baseline Learner

Vibhorpandhare, Xiaodong Jia, Jay Lee
2021 Proceedings of the Annual Conference of the Prognostics and Health Management Society, PHM  
To demonstrate the ability of federated learning to enhance the robustness and reliability of PHM models, this paper proposes a novel federated Gaussian Mixture Model (GMM) algorithm to build universal  ...  Federated collaborative learning can serve as a catalyst for the adaptation of business models based on the servitization of assets in the era of Industry 4.0.  ...  Federated training of the global baseline GMM model in the global server is performed by the exchange of model parameters with each client using the local baseline data as described in the proposed algorithm  ... 
doi:10.36001/phmconf.2021.v13i1.2989 fatcat:yotbx3xk2jff3iak5dpvxfn32a

Federated Learning for Open Banking [article]

Guodong Long, Yue Tan, Jing Jiang, Chengqi Zhang
2021 arXiv   pre-print
The most attractive aspect of federated learning is its ability to decompose model training into a centralized server and distributed nodes without collecting private data.  ...  This is a just-in-time technology that can learn intelligent models in a decentralized training manner.  ...  [9] proposes a mixture model of global and local models in a federated setting. The local model can preserve the personalized information and the global model provides common information.  ... 
arXiv:2108.10749v1 fatcat:qbn5q6poxbdyfnktnt6svavbxm

Model Fusion with Kullback–Leibler Divergence [article]

Sebastian Claici, Mikhail Yurochkin, Soumya Ghosh, Justin Solomon
2020 arXiv   pre-print
Our algorithm is easy to describe and implement, efficient, and competitive with state-of-the-art on motion capture analysis, topic modeling, and federated learning of Bayesian neural networks.  ...  The components of the dataset posteriors are assigned to the proposed global model components by solving a regularized variant of the assignment problem.  ...  FA9550-19-1-031, of National Science Foundation grant IIS-1838071, from the MIT-IBM Watson AI Laboratory, from the Toyota-CSAIL Joint Research Center, from a gift from Adobe Systems, and from the Skoltech-MIT  ... 
arXiv:2007.06168v1 fatcat:krycdvisjbfhrncolbckv5ggzi

Communication-Efficient Federated Learning over MIMO Multiple Access Channels [article]

Yo-Seb Jeon, Mohammad Mohammadi Amiri, Namyoon Lee
2022 arXiv   pre-print
We then analyze the reconstruction error of the proposed algorithm and its impact on the convergence rate of federated learning.  ...  Communication efficiency is of importance for wireless federated learning systems.  ...  In federated learning, each wireless device updates a local model based on its local training data set and then sends the local model update (e.g., a gradient vector of the local model) to the PS.  ... 
arXiv:2206.05723v1 fatcat:pxvgotp6jvgodaders3grxv7vm

Delta-DAGMM: A Free Rider Attack Detection Model in Horizontal Federated Learning

Hai Huang, Borong Zhang, Yinggang Sun, Chao Ma, Jiaxing Qu, Van-Hau Nguyen
2022 Security and Communication Networks  
Federated learning is a machine learning framework proposed in recent years. In horizontal federated learning, multiple participants cooperate to train and obtain a common final model.  ...  Participants only need to transmit the local updated model instead of local datasets.  ...  Yu Haining for his guidance of this paper and the National Natural Science Foundation of China (62172123) for its support.  ... 
doi:10.1155/2022/8928790 fatcat:s5veaktnfndrjksvqgazisyzzy

Decentralised Person Re-Identification with Selective Knowledge Aggregation [article]

Shitong Sun, Guile Wu, Shaogang Gong
2021 arXiv   pre-print
the global model aggregation in federated Re-ID learning.  ...  To resolve this problem, two recent works have introduced decentralised (federated) Re-ID learning for constructing a globally generalised model (server)without any direct access to local training data  ...  Federated Learning aims to learn a global model with the collaboration of multiple lo- cal models.  ... 
arXiv:2110.11384v1 fatcat:ihe4purgebd5rf5sp56hrjymfa

Resource-Constrained Federated Learning with Heterogeneous Labels and Models [article]

Gautham Krishna Gudur, Bala Shyamala Balaji, Satheesh K. Perepu
2020 arXiv   pre-print
Particularly, on-device federated learning is an active area of research, however, there are a variety of challenges in addressing statistical (non-IID data) and model heterogeneities.  ...  framework by experimenting with federated learning and inference across different iterations on a Raspberry Pi 2, a single-board computing platform.  ...  Mohri et al. propose Agnostic Federated Learning [18] , which addresses about handling any target data distribution formed by a mixture of client distributions.  ... 
arXiv:2011.03206v1 fatcat:yk4hxhyd7vahfb7qvfi44id2ji

No Fear of Heterogeneity: Classifier Calibration for Federated Learning with Non-IID Data [article]

Mi Luo, Fei Chen, Dapeng Hu, Yifan Zhang, Jian Liang, Jiashi Feng
2021 arXiv   pre-print
A central challenge in training classification models in the real-world federated system is learning with non-IID data.  ...  from an approximated gaussian mixture model.  ...  Acknowledgement We would like to thank the anonymous reviewers for their insightful comments and suggestions.  ... 
arXiv:2106.05001v2 fatcat:fmbnjqpmdbc6dj3mlc6usg3nhe
« Previous Showing results 1 — 15 out of 29,333 results