Enzo Zhang zifan wu haihui wang Ryan Anderson
All attributes of each row data for Federated Machine Learning come from more than one edge provider. They must be combined as input of Federated ML model in AI Edge case. Each side has no ability to hold all required data. Edge side has much pretty limitation for edge computing because of variety of conditions.
The application must cover edges more than one to collect and exchange data between edges. Meanwhile a method must be applied to make sure the communication secure enough for all participants to protect data.
Here are some AI edge characters as following.
- Deploy more than one edge
- No more extra storage space
- A few computing units
- Deal with data locally
- Network ability
- Share data each others
- Work together with FML framework
- Data security
Application groups data from edges base on security encryption method. Data privacy is still going to be managed by Federated Machine Learning framework. Data as input of machine learning model was hold in all edges after exchange in secure way, then learning process execute. Federated Machine Learning model with required data was used in edge side for next process.
Many key components are required to make the case works well and machine learning run success.
- Data transfering and grouping
- Data collect and distribute
- Federated Machine Learning framework
- Secure encryption algorithm
Federated Machine Learning framework support AI Edge application to run in transparent way with high functionality.