Hello and Welcome! I am Phillip Christoffersen, a doctoral student at the Massachusetts Institute of Technology, advised by Dylan Hadfield-Menell at the Algorithmic Alignment Group.

Broadly speaking, I am interested in the intersection of AI Alignment and Reinforcement Learning (RL). Currently, I am conducting research in multi-agent RL, neurosymbolic RL, and more generally in formalizing conceptual questions in AI alignment as they apply to RL. However, my interests remain broad, and I am interested in any approach that gets us closer to reliably performant RL systems that demonstrably do what we genuinely want them to.

Before coming to MIT, I was an undergraduate at the University of Toronto (specifically in the Department of Computer Science), and a student researcher at the Vector Institute, where I was advised by Sheila McIlraith.

Moreover, in my spare time, I like to play piano, compose music, read, code, and learn about areas of research outside of my own, in order to maintain a broad perspective. If you are interested in any of these things, or just wanna chat, feel free to email me, otherwise contact me on LinkedIn or Twitter!