Ilmu Komputer & AI editorial
Open AccessOA2022
Federated Learning: Collaborative Machine Learning without Centralized Training Data
Federated learning enables training machine learning models across decentralized devices without sharing raw data, addressing privacy, security, and access rights. This digest extracts the core IMRAD elements from the source paper.
V. Abhishek; Binny S; T. Johan; Nithin Raj; Vishal Thomasยท International Journal of Engineering Technology and Management Sciencesยท 2022ยท DOI 10.46647/ijetms.2022.v06i05.052
The core problem
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.
Innovation
The source paper describes federated learning as a technique that trains an algorithm across multiple decentralized edge devices or servers holding local data samples, without exchanging them. The methodology involves collaborative model training where actors contribute to a shared model without sharing raw data. The paper contrasts this with centralized machine learning, where all data is uploaded to a single server, and with traditional decentralized approaches that assume uniform data distribution. The key methodological innovation is the ability to address data privacy, security, access rights, and heterogeneous data challenges by keeping data localized.
Introduction
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.
Why it matters
The discussion highlights that federated learning offers a paradigm shift from centralized machine learning by allowing collaborative model training without data centralization. This is particularly beneficial for scenarios where data privacy and security are paramount, such as in defence and healthcare. The paper notes that federated learning can handle heterogeneous data distributions, unlike traditional decentralized methods that assume uniform distribution. The authors suggest that federated learning has broad applicability and can mitigate risks associated with data sharing while enabling robust model development.
Who should read this
Pembaca Research Digest dan praktisi rumpun ini.
Opening member contentโฆ