Privacy Teaching Series: What is Static vs Dynamic Federated Learning?

[em-blue]

This has been asked a lot frequently - @cereallarceny I think it'd be good to explain this in a quick blog post to save you time re-explaining. Or perhaps 'model centric' vs 'data centric' or however you'd like to classify. Thoughts?

[cereallarceny]

We will get around to this, yes. I’ll have this in time. In short, here’s the difference:

Federated learning: training or inference of a model on data where the model and data exist in separate locations Static federated learning: federated learning where the model is hosted in a central location and the data is at the edge (in a web browser or on a mobile device) Dynamic federated learning: federated learning where the data is hosted in a central location and the model is at the edge (created by a data scientist)

[Kritikalcoder]

I'd love to help with this

[cereallarceny]

As would I! @Kritikalcoder @em-blue

[em-blue]

Thanks @cereallarceny I'll be marking this as closed as it's covered in your latest :-) https://blog.openmined.org/announcing-new-libraries-for-fl-on-web-and-mobile/