Mathematics Colloquium

AI / Scientific Computing: Alloys for Modeling and Control of Complex Systems

Speaker: Petros Koumoutsakos, Harvard

Location: Warren Weaver Hall 1302

Date: Monday, February 13, 2023, 3:45 p.m.


Over the last thirty years, we have experienced more than a billion-fold increase in hardware capabilities and a dizzying pace of acquiring and transmitting massive amounts of data. Scientific Computing and Artificial Intelligence (AI) have been key beneficiaries of these advances. This talk will offer a perspective on forming algorithmic alloys of AI and simulations to predict and control complex systems. 

The applications have an emphasis on fluid mechanics. Alloys will include algorithms for learning the Effective Dynamics (LED) of complex systems and a fusion of multi-agent reinforcement learning and scientific computing (SciMARL) for modeling and controlling turbulent flows. The talk will present successes and failures and argue that the proper fusion of domain knowledge and AI expertise is essential for advancing scientific frontiers.