Anyone used Prefect to do federated online learning? Having a centralized model, do prediction on individual devices without seeing the data, get feedback from the error of those predictions and then send back something like the gradient to update the centralized model? Any resources to read more about this issue in particular?
07/07/2020, 3:48 PM
Offhand I’m not aware if anyone is but online learning is an old research interest of mine so I’m curious to see if anyone responds!
07/07/2020, 3:58 PM
I found Pysyft and trying to figure out if I can combine Prefect and it or not 🙂 (maybe Prefect can handle higher level task execution and Pysyft can handle updating the model ) https://github.com/OpenMined/PySyft