What must machine learning models memorize?

October 3rd, 2021

by Gavin Brown -- Machine learning models serve a variety of purposes in today's society, such as identifying faces in images or helping customers via a chatbot. Many models are trained on vast amounts of (possibly sensitive) data. The models themselves are large: OpenAI's celebrated GPT-3 model for generating text has 175... More

Confidentiality and the Census (Part 2)

June 29th, 2021

by Aloni Cohen Be sure to check out part 1 of this blog post here.   Data Quality Confidentiality is critical for a useful, high quality Census. Perfect confidentiality is easy to achieve if we don't care about doing anything with the Census responses. Just erase them. Or don't collect them at all. But... More

Confidentiality and the Census

June 21st, 2021

By Aloni Cohen   A political fight is brewing around the 2020 Census. No, not about the citizenship question. This fight is about confidentiality—what's the best way to publish useful statistics without disclosing individual census responses? At the center of this fight is a lawsuit between the state of Alabama and the... More