Why We Need to See Inside AI’s Black Box

This article was originally written by me for The Conversation, an online publication covering the latest research. It appeared there on May 26, 2023. For some people, the term “black box” brings to mind the recording devices in airplanes that are valuable for postmortem analyses if the unthinkable happens. For others it evokes small, minimally […]

The Caprice of Algorithms

When we design algorithms or implement them into computing systems, we rarely think about the policies that they instantiate. We rarely think beyond trite generalizations, of the countless users who may be at the receiving end of the vagaries of our algorithms. If our project becomes immensely successful, then we will count our users in […]

Three algorithms to live by: LSTM, FedAverage, C-W

Citation We come across many algorithms in our education and work. Here I look at three relatively recent ones from the area of Machine Learning (ML), and more specifically from my vantage point of reliability and security of ML. There have been reams written on each, at various levels of technical depth. So obviously I […]