Ieee federated Machine Learning White Paper
Download 0.56 Mb. View original pdf
|
e-Iraq estra.ar.en
- Navigate this page:
- secure and privacy-preserving techniques
| 7 IEEE SA 1.3. WHY IEEE FEDERATED MACHINE LEARNING (IEEE STD 3652.1-2020) In response to the urgent need fora Federated Machine Learning technology framework, the IEEE P Working Group came together to create an architectural framework and application guidelines for federated machine learning (FML), which includes the following A description and definition of federated machine learning The categories of federated machine learning technologies and the application scenarios to which each category applies A set of measures concerning the performance evaluation criteria for federated machine learning Associated features of federated machine learning that fulfill different regulatory requirements (see IEEE Std 3652.1-2020 [1]). This white paper does not detail the technical content of the guide rather, the white paper illustrates the need of such a guide by showcasing a variety of use cases of the FML frameworks defined in the guide. By doing so, the hope is that the white paper will provide readers with a brief overview of the technological landscape of FML as well as underlying principles concerning the implementation of the FML framework in real-life applications. 2. FRAMEWORK AND CATEGORIZATION OF FEDERATED MACHINE LEARNING Federated machine learning is ab distributed machine-learning framework that enables multiple participants to collaboratively train and use a machine learning model fora given task, e.g., classification, prediction, and recommendation. Within this framework, all raw data owned by different participants are protected by secure and privacy-preserving techniques, which prevent the data from being tampered with and disclosed by other participants or reverse-engineered by other participants. FML, as a machine-learning framework, first concerns the performance of the learned models. It is expected that any sound FML methods maintain performance that is very close to that of the model built when data from multiple participants were put together in one location. Second, due to the distributed learning nature of FML, the learning Download 0.56 Mb. Share with your friends:
|