FACT: Federated Adversarial Cross Training

1 code implementation • 1 Jun 2023 • Stefan Schrod, Jonas Lippl, Andreas Schäfer, Michael Altenbuchinger

We propose Federated Adversarial Cross Training (FACT), which uses the implicit domain differences between source clients to identify domain shifts in the target domain.

Federated Learning Multi-Source Unsupervised Domain Adaptation +2