Teaching and Service

PH 290 · Fall 2026

We start from the fundamentals of generative AI, tokenization, transformer attention and pre-training, then move through in-context learning, fine-tuning, retrieval-augmented generation, multimodal language models and model evaluation. Students apply these methods to public health problems in final class projects.

Syllabus (PDF) ↗

PH C240C / STAT C245C · Fall 2020, 2021, 2022, 2023; Spring 2025

An overview of, and hands-on experience with, the machine learning methods and biostatistical models used in the healthcare system and in medical research. Topics run from supervised learning (GLMs, support vector machines, metric learning, tree-based and shrinkage-based approaches) and semi-supervised learning to deep learning and neural networks, adaptive experiments, reinforcement learning and multi-armed bandits, causal inference and resampling-based inference. Applications are covered alongside the methods.

Listed through Spring 2025 under its former title, Computational Statistics with Applications in Biology and Medicine.

PH 252D · Spring 2022

A general framework for causal inference built on directed acyclic graphs, nonparametric structural equation models and counterfactuals. The course introduces marginal structural models and the estimation of causal effects by inverse probability of treatment weighting, G-computation and targeted maximum likelihood. Students define and implement a research question of their own across two presentations.

PH C240A / STAT C245A · Fall 2019; Spring 2021

Major topics in general statistical theory, with a focus on the methods used in epidemiology and a framework for understanding the properties of both the common ones and the more advanced. The emphasis is on estimation in nonparametric models: maximum likelihood and loss-based estimation, asymptotic linearity and normality, the delta method, bootstrapping, machine learning and targeted maximum likelihood estimation. Broad concepts come first, with implementation in R alongside them.