| |
Releasing SQL Queries with Optimal Error under Differential Privacy
|
Speaker (s):
DONG Wei
PhD Candidate,
Computer Science and Engineering
Hong Kong University of Science and Technology
|
|
Date:
Time:
Venue:
|
|
21 December 2022, Wednesday
10:00am - 11:15am
This is a virtual seminar. Please register by
16 December 2022, the meeting link will be sent to those who have registered on the following day.
We look forward to seeing you at this research seminar.
|
|
About the Talk
Differential privacy (DP) has become the standard notion for private data release, due to its strong protection of individual information. There are two main challenges. First, informally speaking, DP requires indistinguishability of the query results whether any particular individual's data is included or not in the database, which is hard to achieve when one individual can bring a large or even unbounded contribution to the query result. Second, for the privacy goal, some noise must be introduced, but given a query, what is the optimal error that one DP mechanism can achieve? In this talk, the speaker will give a selective overview of his recent research in solving above two challenges in SQL queries answering.
About the Speaker
Mr Wei Dong is now a Ph.D. candidate in the Department of Computer Science and Engineering at the Hong Kong University of Science and Technology. His general areas of interest include database theory and algorithms, data security and privacy, and statistics. His research has been recognized by the academic community and appeared in top conferences, such as SIGMOD, PODS, NeuIPS, CCS, and KDD. He received the Best Paper Award in SIGMOD 2022.
He is a tenure-track faculty candidate for the Artificial Intelligence & Data Science, Data Management & Analytics.
|
