FedDis
We use federated learning to train collaboratively an unsupervised neural network on multiple institutes in a privacy-aware manner with the goal of segmenting brain pathology on MRI scans. We show that the federated paradigm offeres an implicit way to disentangle the shape and appearance of brain scans, learning representations that are robust to the domain shifts of the different institutes.
Check our publications below for more details.
Cosmin I. Bercea
Doctoral Researcher
My research is focused on interpretable machine learning for anomaly detection.
