Differential Privacy for the 2020 U.S. Census: Can We Make Data Both Private and Useful? (Part 1)
Harvard Data Science Review Podcast
00:00
What Is Differential Privacy?
The citizenship question, of having that question on the census of our u.S. citizen or not, was so controversial? Would that be because you would not get an accurate acount of the number of people living in the united states if it's not private? The idea really is that if we were to compute that function on the data base with my information inside the data base versus my information outside of the data base, then result should not be different by too much. And therefore, if that cold happen, this function will faithfully capture the information it intends to capture. But just by looking at this function alone, anyone would not be to tell whether i, as a person, is inside
Transcript
Play full episode
