Jupyter Notebook Tutorial – Privacy-Preserving Deep Learning in PyTorch

[williamtrask]

https://github.com/OpenMined/PySyft/tree/master/examples/tutorials

[williamtrask]

Based on PySyft: https://github.com/OpenMined/PySyft

Paper: https://arxiv.org/abs/1811.04017.pdf

[brandoncarl]

It's fantastic to see great research combined with great tutorials – bravo. You all have bitten off an incredibly hard problem with pragmatism and tenacity. Keep up the great work.

[williamtrask → brandoncarl]

Thank you!

[Sol-]

My first thought was that this complements differentially private systems nicely, but seems like it was your intention to use it as a building block for things like that anyway. Very cool work! I think it's really important to have such secure primitives that are easy to build upon, maybe secure multiparty computation and such will finally become more widespread.

[williamtrask → Sol-]

Thank you so much!