Dr Shahrokh Daijavad

Title of the Keynote Talk

Enterprise Context Federated Learning: Challenges and Approaches

Biography

Dr Shahrokh Daijavad is a Distinguished Research Staff Member in the Next Generation Computing Systems Department at IBM Almaden Research Center. He received B.Eng. (Summa Cum Laude) and Ph.D. degrees in electrical engineering from McMaster University, Canada in 1983 and 1986, respectively. After a year of postdoctoral fellowship at the University of California at Berkeley, he joined IBM T. J. Watson Research Center in Yorktown Heights and worked there from 1987 to 2001. From 2001 to 2003, he was the CTO of a Japanese Internet start-up company. He returned to IBM at the end of 2003 and has been the software lead in the Next Generation Computing Systems group since then, working at the Watson Research Center from 2003 to 2013 and moving to the Almaden Research Center in 2013. Since 2014, after co-leading the GTO topic “Ad Hoc Data Infrastructure and Data”, his focus has been on IoT technologies and solutions, with particular emphasis on “edge computing”. In October 2017, he was the panel moderator on “Enabling Technologies for Edge Computing” at the ACM/IEEE Symposium on Edge Computing (http://acm-ieee-sec.org/2017/index.html)

Dr Jakub Konečný

Title of the Keynote Talk

Federated Learning from Research to Practice

Abstract

Federated Learning enables mobile devices to collaboratively learn a shared prediction model while keeping all the training data on device, decoupling the ability to do machine learning from the need to store the data in the cloud. In this talk, I will discuss: (1) how federated learning differs from more traditional machine learning paradigms; (2) practical algorithms for federated learning that address the unique challenges of this setting; (3) extensions to federated learning, including differential privacy, secure aggregation, communication and computational efficiency, and (4) an overview of federated learning applications and systems at Google.

Biography

Dr Jakub Konečný is a Research Scientist at Google where he works on Federated Learning and other privacy-preserving techniques for data analysis. Previously, he completed his PhD at the University of Edinburgh, focusing on distributed optimization for Machine Learning, under supervision of Peter Richtárik with support through Google PhD Fellowship.