Fairness in machine learning has been increasingly more important in recent years due to social responsibility.
Even if many algorithms have been proposed to address fairness concerns by introducing constraints or regularization in the conventional setting, only a few can deal with federated setting, where local private data is prohibited from sharing across users and central servers,
and estimation in the population level can be painfully difficult.
In this paper, we propose a robust and efficient algorithm that ensures R'enyi-based fairness and can be well suited in the federated setting.
The proposed algorithm aggregates local statistics required by the R'enyi correlation with a simple yet robust estimator, under the challenge of federated privacy constraints.
Our analysis provides decent concentration between such local aggregated empirical estimation and global population in
-
Notifications
You must be signed in to change notification settings - Fork 0
AllenMa97/Federated-Renyi
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
No description, website, or topics provided.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published