Virtual Meetup: Privacy-Preserving ML

While in-person events are on hold, we'd like to invite you to this virtual meetup organized by the Data Council Barcelona meetup group - just RSVP here to get joining details**

Talk 1: Privacy-Preserving Machine Learning

Speaker: Romeo Kienzler, Chief Data Scientist at IBM Center for Open Source Data and AI Technologies (CODAIT)
https://www.linkedin.com/in/romeo-kienzler-089b4557

Abstract:
"Data privacy is a huge concern and often prevents ML and AI project from flourishing. In this talk we’ll introduce you to federated learning and homomorphic encryption. After we’ve covered the theoretical aspects we’ll see how they can be used in practice. We conclude with an outlook on the future of these technologies."

Talk 2: What is Collaborative Learning?

Speaker:
Morten Dahl, PhD, Head of Research at Cape Privacy
https://www.linkedin.com/in/mortendahlcs/

Abstract:
"It's well known that machine learning models reflect the quality and quantity of the data you input (Garbage in, Garbage out!). If companies want to move forward and rival the AI capabilities of larger data collectors, data partnerships, data sharing and collaborative machine learning are powerful solutions. Currently, these practices are underutilized as a way to create better models. In this talk, we'll discuss how to begin planning for collaborative learning and data sharing in a privacy-aware and secure manner. What security threats or sensitive data concerns should you consider and incorporate in the collaboration? What level of trust is needed between the different data partners and how can that trust be best managed? How could collaborative learning create a mutually beneficial experience for your organization and customers? We'll address the challenges and benefits of collaborative learning and highlight advanced privacy and security techniques designed for machine learning at scale.In this talk, we’ll discuss how to begin planning for collaborative learning and data sharing in a privacy-aware and secure manner, and overall become more mature with respect to privacy both as a data science team and as a company."