Hi, I am William. This is my personal blog where I write about my interests (which are many) and scratch my childhood itch of being a writer.
Research Interests
My current research interests focus on how multiple agents learn to act well over time: the mechanics of reinforcement learning, the strategic reasoning captured by game theory, and what it takes to develop genuine skill in hard domains. A common theme running through all of this is representation learning — the question of what an agent (or a person) learns to notice and abstract from experience, and how the quality of those internal representations shapes everything downstream.
These aren’t just for technical interests; I find myself thinking about skill acquisition, deliberate practice, and the architecture of expertise in my own learning and skills acquisition — like in skateboarding and language learning — as much as I do about algorithms.
Running alongside all of this is my personal interest in mathematical topics such as topology, graph/category/group theory, and geometry — drawn in by the way they reveal deep structure, such as symmetries, beneath the surface of everyday things.
Publications
Teo, W. “A natural language processing approach to improve demand forecasting in long supply chains.” Master of Engineering Thesis, MIT, 2020.
A deep learning model (NEMO) that uses NLP signals from public text to forecast commodity demand in long supply chains, where information sharing across firms is rarely feasible.