Federated Learning: Collaborative Machine Learning without Centralized Training Data

By Abhishek V A, Binny S, Johan T R, Nithin Raj, Vishal Thomas. 2022

Federated learning (also known as collaborative learning) is a machine learning technique that trains an algorithm without transferring data samples across numerous decentralized edge devices or servers. This strategy differs from standard centralized machine learning techniques in which all local datasets are uploaded to a single server, as well as more traditional decentralized alternatives, which frequently presume that local data samples are uniformly distributed. Federated learning allows several actors to collaborate on the development of a single, robust machine learning model without sharing data, allowing crucial issues such as data privacy, data security, data access rights, and access to heterogeneous data to be addressed. Defence, telecommunications, internet of things, and pharmaceutical industries are just a few of the sectors where it has applications.

Subfields: Federated & Distributed Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Federated & Distributed Learning

Citations in the corpus

Advances and Open Problems in Federated Learning (2019, background): “Since the term federated learning was initially introduced with an emphasis on mobile and edge device applications [289, 287], interest in applying FL to other applications has greatly increased, including some which might involve only a small number of relatively reliable clients, for example multi”
Generative Models for Effective ML on Private, Decentralized Datasets (2019, neutral): “In other settings—including federated learning (FL) (McMahan & Ramage, 2017; ar X iv :1 91 1.”
FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantization (2019, background): “Indeed, there are multiple factors that determine whether a device can participate in the training [McMahan and Ramage, 2017]: a device should be available in the reachable range of the base station; a device should be idle, plugged in and connected to a free wireless network during the training; et”
Private Aggregation from Fewer Anonymous Messages (2019, methodology): “For learning algorithms based on gradient descent, a secure aggregation primitive can be used to compute global weight updates without compromising privacy [23, 24].”
Distributed Training with Heterogeneous Data: Bridging Median- and Mean-Based Algorithms (2019, methodology): “[15], McMahan and Ramage [17] for collaborative training of machine learning models without transmitting users’ data.”
Decentralized Privacy Using Blockchain-Enabled Federated Learning in Fog Computing (2020, background): “The federated learning [14], which is a new-merging technology, can achieve efﬁciency communication by exchanging updates between local models and global models [2].”
Multi-Objective Evolutionary Federated Learning (2018, background): “Federated learning [21] is an emerging decentralized privacy-protection training technology that enables client edges to learn a shared global model without uploading their private local data to a central server.”
A Learning-Based Incentive Mechanism for Federated Learning (2020, background): “Since an inception by Google [4], federated learning has attracted great attentions from both academia and industry.”
FLAME: Taming Backdoors in Federated Learning (2022, background): “1, we generally focus on FedAvg in this work as it is commonly applied in FL [21, 28, 39, 44, 47, 50, 54] and related work on backdoor attacks [7, 22, 51, 57, 59].”
A Crowdsourcing Framework for On-Device Federated Learning (2019, methodology): “Setting a different motivation, FL recently has attracted an increasing interest [7], [11], [12], [13], [15], [22] in which collaboration of the number of devices with non-i.”
Communication-Efficient Federated Learning for Wireless Edge Intelligence in IoT (2020, methodology): “To address the issue of distributed ML, the concept of Federated Learning (FL) [2] – collaboratively training a model across devices without sharing their data – was introduced.”
Communication-efficient federated learning (2021, unknown): “, Google (7)], speaker and command recognition [e.”
A first look into the carbon footprint of federated learning (2020, methodology): “While FL is still a maturing technology, it is already being used by millions of users on a daily basis; for example, Google uses FL to train models for: predictive keyboard, device setting recommendation, and hot keyword personalization on phones (McMahan and Ramage, 2017).”