Federated Learning with DL4J and Gluon

This virtual meetup starts at 10 am in San Francisco, 1 pm in the East Coast, 6 pm in London, 10:30 pm Bangalore, India (Time zone converter -
https://www.timeanddate.com/worldclock/converter.html?iso=20190814T170000&p1=283&p2=179&p3=136&p4=166&p5=438&p6=33&p7=248)

Use the following link to find the live event on YouTube:
https://www.youtube.com/channel/UC1PncmBLZMqlodEt7atfFuw

Abstract:
We will demonstrate an end-to-end application in which Java-based deep-learning tools are used on (desktop/mobile/embedded) clients to analyze and train data.
A domain-specific model is maintained by a server and enhanced by updates being sent from different clients. In many cases where machine learning can provide benefits, the input data is privacy-sensitive. Although more input data will lead to a more accurate model on the back end, it is often not desirable to share the raw input data. In this case, federated learning helps by analyzing the data, retraining the model on the client, and sending the resulting changes back to the server.
The presentation shows how this can be done using the SkyMind deeplearning4j library, and Gluon Mobile tools.

Speaker: Johan Vos, CTO, Gluon
Johan Vos started to work with Java in 1995. He was part of the Blackdown team, porting Java to Linux. His main focus is on end-to-end Java, combining back-end systems and mobile/embedded devices.
He received a Duke Choice award in 2014 for his work on javafx on mobile. In 2015, he co-founded Gluon, which allows enterprises to create (mobile) Java Client applications leveraging their existing backend infrastructure. Gluon received a Duke Choice award in 2015.
Johan is a Java Champion, a member of the BeJUG steering group, the Devoxx steering group and he is a JCP member. He is the lead author of the Pro JavaFX 8 book, and he has been a speaker at numerous conferences on Java.
@johanvos