Cyril Zhang

À propos

I’m broadly interested in theory and algorithms for machine learning. My work has weaved through these topics in particular:

• Temporality. Learning and decision-making in dynamical systems. Algorithm design for {non-stationary, stateful, misspecified, adversarial} environments. Principles of {continual, transfer} learning.
• Optimization. Computational and statistical lenses on the trajectory of learning. Practical and reusable algorithmic tools for training large-scale models, from linear regression to modern neural nets.
• Language. Modeling discrete sequences with combinatorial structure, memory, and intent. My favorite source of subproblems in the pursuit of understanding cognition.

I completed my Ph.D. in Computer Science at Princeton, under the supervision of Prof. Elad Hazan. For the latter half of that wonderful adventure, I was a student researcher at Google AI. Before that, I completed a B.S. in Computer Science at Yale.