Membership Inference Attack in Federated Learning

Just a prototype now.

And welcome to post issues to reveal hidden faults, since there were many unjustified hypotheses during coding.

Acknowledge

• Thanks to Mihir Khandekar and Zing22, whose patience and replies keep this project going. And it's my pleasure to submit issues to point out some marginal flaws on the repositories they maintain.
  • ML Privacy Meter
  • tf-fed-demo
  • What I am doing can be regarded as an attempt to combine their contributions.