AI-enabled Systems: Safe, Explainable, and Trustworthy (ASSET) Center

Jiayuan Mao aims to build intelligent agents that continually learn, reason, and plan: agents that can answer queries, infer human intentions, and pursue long-horizon goals spanning hours to days. In this talk, she will describe a general framework for learning and reasoning with neuro-symbolic concepts. Drawing inspirations from theories and studies in cognitive science, neuro-symbolic […]

The recent autonomous solution to the Navier-Stokes millennium problem was an extraordinary breakthrough. Yet, equally remarkable is how the global community accepted this solution without traditional, years-long peer review. This speed was possible because the proof included an autoformalized Lean certificate, allowing it to be mechanically verified. Lean was the foundational trust substrate beneath the […]

Power grids must be managed at greater speed, scale, and physical fidelity to enable decarbonization and improve resilience. Methods from machine learning (ML) have the potential to play an important role by providing fast, scalable approximations for foundational power system optimization problems, but often struggle to enforce hard physical constraints, maintain robustness in the face […]

PDeep learning models have become remarkably powerful – but often operate as black boxes. In this talk, I will share a series of work from my lab on how to make these systems more transparent, reliable, and trustworthy. I’ll highlight three research directions to bring interpretability into deep learning: (1) automated tools that reveal what […]