ML

Intro to Sampling using Hamiltonian Monte-Carlo

Tutorial on HMC, which uses Hamiltonian dynamics to efficiently sample from high-dimensional distributions

Reading Group: Adversarial Spheres

Notes on the 2018 paper Adversarial Spheres, which shows how adversarial examples may be inevitable given an imperfect, high-dimensional classifier

Random matrix theory via undergrad physics

Uses basic results from undergraduate quantum and statistical mechanics classes to explore the Gaussian Orthogonal Ensemble, one of the canonical probability distributions studied in random matrix theory.