I’m an incoming PhD student in the machine learning department at Carnegie Mellon University. Currently, I’m the lead research analyst at Stanford Law School’s Regulation, Evaluation, and Governance Lab (RegLab).
On the side, I co-host the Increments podcast, where Vaden Masrani and I yell at each other and others about philosophy and science. Sometimes I also write about these topics. I often distract myself with good books and interesting online courses.
You can contact me at benchugg \(\alpha\tau\) stanford.edu.
My primary interests at the moment involve developing robust sequential decision-making algorithms, particularly through a non-parametric and non-asymptotic lens, that are as assumption free as possible. I’ve previously worked on problems in graph theory, combinatorial optimization, and stochastic reaction networks.
Ideally, I try and straddle the borderlands between theory and practice with a bias towards the former, but apparently the results have high variance because I’ve spent time doing doing both pure theory and pure application.
Prior to Stanford, I was a master’s student at the Mathematical Institute at Oxford, where I studied topics at the intersection of spectral graph theory and geometry. Here is my thesis on the subject.
The podcast is mostly an attempt to break out of the academic echochamber and talk about things other than limit theorems and algorithmic performance. We have a few beers and chat about anything from moral philosophy and thought experiments, to the philosophy of logic, climate change, free will, dualism, and social media.
We also have an ongoing series on Conjectures and Refutations by Karl Popper.