3 ms·

There is so much hype around federated learning but often the hard and insurmountable part of this is federated labeling. For example for your cancer use case,

There is so much hype around federated learning but often the hard and insurmountable part of this is federated labeling. For example for your cancer use case, you have to convince multiple hospitals to feed the system labels and this is a very very tall ask. For healthcare it’s also not clear how to get a regulatory clearance if you can’t actually test the performance of the federated deployments. So while federated learning solves some problems generated by an unwillingness to share data, it doesn’t solve all of them. Describe the use cases of your product carefully.

niclane7 4y ago
Regarding federated labeling, you might be interested in some recent prototypes built on Flower that use forms of self supervised learning. By combining SSL with federated learning we can start to leverage unlabeled data and this will be a big deal once it becomes common place. I'd suggest looking at these two research papers that build on Flower and include members of the Flower team as authors: https://arxiv.org/abs/2207.01975 https://arxiv.org/abs/2207.01975 https://arxiv.org/abs/2204.02804 https://arxiv.org/abs/2204.02804