Widely Applied Math Seminar
Thursday, December 5
3pm | Pierce 209


Building a Science Assistant
Michael Brenner
Catalyst Professor of Applied Mathematics and Applied Physics and of Physics; Area Chair, Applied Mathematics

Abstract: I will discuss recent efforts to build a science assistant, with the goal of using LLM technology to accelerate the practice of science. In particular we focus on the manual creation of software to support computational experiments, using an AI system that creates expert-level scientific software whose goal is to maximize a quality metric. The system uses a Large Language Model  and a variant of Monte Carlo Tree Search  to systematically improve the quality metric and intelligently navigate the large space of possible solutions. I will demonstrate the system on various types of computational problems, showing how it achieves expert-level results when it explores and integrates complex research ideas from external sources.   Examples range from analysis of single cell RNA sequencing data, where the method discovers 40 novel methods for single-cell data analysis to  epidemiology (where it generated 14 models that outperformed the CDC ensemble and all other individual models for forecasting COVID-19 hospitalizations) to applied math problems of various types that might be of interest to this audience, ranging from very difficult problems in applied math 201 to solving nonlinear PDEs etc.

I'm also quite interested in finding use cases where this tool could be useful for people's research.

Paper https://arxiv.org/pdf/2509.06503

Bio: Michael Brenner is the Catalyst Professor of Applied Mathematics and Applied Physics, and Harvard College Professor at the Harvard School of Engineering and Applied Sciences. He developed the popular Harvard class, "Science and Cooking: From Haute Cuisine to the Science of Soft Matter," with his colleague David Weitz and chef Ferran Adrià. His research uses mathematics to examine a wide variety of problems in science and engineering, ranging from understanding the shapes of bird beaks, whale flippers and fungal spores, to finding the principles for designing materials that can assemble themselves, to answering ordinary questions about daily life, such as why a droplet of fluid splashes when it collides with a solid surface.