AI & Dynamics Group · University of Rochester
The dynamics of learning,
§ Research
All projects →
ML for Dynamics
→
Learning of dynamics
Forecasting and discovering nonlinear systems with machine learning — reservoir computers, foundation models, hybrid physics-ML methods — and the failure modes of each.
Physics of AI
→
Dynamics of learning
Machine learning as a physical system — training trajectories, learned geometries, phase transitions — and the lens that explains when ML works and when it breaks.
Networks & Emergence
→
Rule of emergence
Simple coupling rules, baroque collective behavior. Synchronization, chimeras, even computation — shaped by network topology, higher-order interactions, and basin geometry.