Thursday, June 25, 2020 - 14:00 to 16:00
SCHOOL OF COMPUTER SCIENCE
The School of Computer Science would like to present…
PhD. Comprehensive Exam by: Samaneh Miri
The rapid explosion in AI has provided the possibility of using aggregated data to produce robust models. However, increasing privacy acts and concerns hinder organizations from taking advantage of available big data because it usually sits in data silos. Federated Learning (FL), as client-side computing, is a promising solution for this challenge. It is a new framework for AI model development that provides collaborative model learning without centralizing training data or exchanging datasets. Privacy by design is one of the primary attractions of FL, which achieves through data minimization, i.e., the raw client data never leaves the device. However, there is still no formal guarantee of privacy, and also recent research shows that attackers can indirectly obtain sensitive information through shared parameters. In addition to providing confidentiality, any privacy-preserving method in the FL setting should be highly communication efficient. In FL, many constraints often co-occur and make the solving problem a multi-dimensional task that includes machine learning, distributed optimization, security, privacy, statistics, compressed sensing, and more. Having a practical FL demands the best possible trade-off between efficiency, privacy, and accuracy which is still an ongoing area of research and needs researchers’ considerations.
Keywords: Collaborative Learning, Federated Learning, On-Device Learning, Privacy, Security.
Date: Thursday, 25th June 2020
Time: 2 pm – 4 pm
Zoom URL: https://zoom.us/j/92999054588?
Internal Reader: Dr. Dima Alhadidi
Internal Reader: Dr. Pooya Moradian Zadeh
External Reader: Dr. Roozbeh Razavi Far
Advisor: Dr. Ziad Kobti, Dr. Saeed Samet
PhD Comprehensive Exam Announcement
5113 Lambton Tower 401 Sunset Ave. Windsor ON, N9B 3P4 (519) 253-3000 Ext. 3716 email@example.com