Deep Federated Learning in Healthcare - AI Quorum | MBZUAI

Описание: Collaborative Learning: From Theory to Practice
Deep Federated Learning in Healthcare
https://mbzuai.ac.ae/the-ai-quorum/

Speaker: Shadi Albarqouni (University Hospital Bonn & Helmholtz Munich)

Overview
We are excited to invite you to the 2022 MBZUAI workshop on Collaborative Learning, organized as part of the AI Quorum series and in partnership with MBZUAI. The field of collaborative learning was conceptualized out of very practical concerns around data privacy and ownership. Since then, there has been tremendous progress in development of new algorithms in theoretical investigations, as well as in emerging real-world applications. Unfortunately, these two branches are growing more separate from each other leading to missed opportunities for progress. The goal of the workshop is to bring disparate communities, all working on different aspects of collaborative learning, together to shape the agenda of future research.

Areas of Focus
• Algorithms and theory – Decentralizing the storage of and computation over the data leads to a wide array of difficulties, many of which are already active topics of research by the community.
• Applications – Real world deployment of CL is still in a nascent stage. Such a transfer of novel technology comes with a host of application-specific challenges. During the workshop, we will shed light on potential challenges and solutions arising during real-world deployment.
• Policy – The rapid growth of interest in CL is a direct product of rising concerns about data governance. Understanding data governance policy will help design better and future-proof CL systems, while understating the technical capabilities and limitations of CL will help design better data governance policies.

Organizing committee:

Michael I. Jordan, UC Berkeley
Sai Praneeth Karimireddy, UC Berkeley
Local committee:
Martin Takáč, MBZUAI
Samuel Horvath, MBZUAI
Eduard Gorbunov, MBZUAI
Local host:
Philip Purnell, MBZUAI