[This post is joint work with Princeton graduate student Changchang Liu and IBM researcher Supriyo Chakraborty. See our paper for full details. — Prateek Mittal ] The tussle between data utility and data privacy Information sharing is important for realizing the vision of a data-driven customization of our environment. Data that were earlier locked up […]
Differential Privacy is Vulnerable to Correlated Data — Introducing Dependent Differential Privacy
August 26, 2016 by
Language necessarily contains human biases, and so will machines trained on language corpora
August 24, 2016 by
I have a new draft paper with Aylin Caliskan-Islam and Joanna Bryson titled Semantics derived automatically from language corpora necessarily contain human biases. We show empirically that natural language necessarily contains human biases, and the paradigm of training machine learning on language corpora means that AI will inevitably imbibe these biases as well. Specifically, we look at […]
Security against Election Hacking – Part 2: Cyberoffense is not the best cyberdefense!
August 18, 2016 by
State and county election officials across the country employ thousands of computers in election administration, most of them are connected (from time to time) to the internet (or exchange data cartridges with machines that are connected). In my previous post I explained how we must audit elections independently of the computers, so we can trust the […]
Security against Election Hacking – Part 1: Software Independence
August 17, 2016 by
There’s been a lot of discussion of whether the November 2016 U.S. election can be hacked. Should the U.S. Government designate all the states’ and counties’ election computers as “critical cyber infrastructure” and prioritize the “cyberdefense” of these systems? Will it make any difference to activate those buzzwords with less than 3 months until the […]
