IEEE 3652.1-2020

Federated learning defines a machine learning framework that allows a collective model to be constructed from data that is distributed across data owners. This guide provides a blueprint for data usage and model building across organizations while meeting applicable privacy, security and regulatory requirements. It defines the architectural framework and application guidelines for federated machine learning, including: 1) description and definition of federated learning, 2) the types of federated learning and the application scenarios to which each type applies, 3) performance evaluation of federated learning, and 4) associated regulatory requirements.

Data privacy and information security pose significant challenges to the big data and artificial intelligence (AI) community as these communities are increasingly under pressure to adhere to regulatory requirements such as the European Union’s General Data Protection Regulation. Many routine operations in big data applications, such as merging user data from various sources in order to build a machine learning model, are considered to be illegal under current regulatory frameworks. The purpose of federated machine learning is to provide a feasible solution that enables machine learning applications to utilize the data in a distributed manner that does not exchange raw data directly and does not allow any party to infer private information of other parties. Federated machine learning is expected to promote and facilitate collaborations among multiple parties, some of which are data source owners, such that user privacy and information security are protected. This guide will promote the use of distributed data sources without violating regulations or ethical considerations.

New IEEE Standard - Active. Federated machine learning defines a machine learning framework that allows a collective model to be constructed from data that is distributed across repositories owned by different organizations or devices. A blueprint for data usage and model building across organizations and devices while meeting applicable privacy, security and regulatory requirements is provided in this guide. It defines the architectural framework and application guidelines for federated machine learning, including description and definition of federated machine learning; the categories federated machine learning and the application scenarios to which each category applies; performance evaluation of federated machine learning; and associated regulatory requirements.