HKBUtube

Distinguished lecture (Hong Kong Baptist University. Dept. of Computer Science)

472 Views 觀看次數

Open to All 公開

00:00

Recommended Videos 影片推薦

Audio 語言

Towards Production Federated Learning Systems

DEPARTMENT:

Department of Computer Science

CONYTRIBUTOR / SPEAKER:

LI, Baochun

RELEASE DATE:

2023

LENGTH:

76 Minutes

VIDEO TYPE:

Scholarly Talks

While there were thousands of papers in the literature in the recent years on federated learning, very few are concerned with its production use in practice. As the (perhaps only) practical paradigm that preserves data privacy when training a shared machine learning model, I expect that federated learning will be widely used in production. But are research results in the literature valid and practical in production federated learning systems? In this talk, I will share our recent experiences with claims in the existing literature along the lines of privacy leakage attacks, and show that their assumptions do not necessarily hold in production systems. I will also introduce more efficient ways to solve the "unlearning" problem, which is necessary due to regulatory constraints in production, such as the GDPR. Our experiments were conducted on Plato, a new open-source federated learning framework that I designed from scratch in the past two years to be as close to production systems as possible, while using a minimum amount of computing resources.

Towards Production Federated Learning Systems

部門:

計算機科學系

講者:

LI, Baochun

發佈日期:

2023

影片長度:

76 分鐘

影片類型:

學術論壇

Towards Production Federated Learning Systems

部门:

计算机科学系

讲者:

LI, Baochun

发布日期:

2023

影片长度:

76 分钟

影片类型:

学术论坛

Full Disclaimer

This video is presented here with the permission of the producers. Any downloading, storage, reproduction, and redistribution are strictly prohibited without the prior permission of the respective producers. Go to Full Disclaimer.