Chemistry Seminar

Modern chemistry produces enormous amounts of data, but insight remains scarce. Mass spectrometry is a striking example: it can capture rich molecular information from complex mixtures, reactions, and materials, yet the resulting data are often too large, too intricate, and too ambiguous to interpret by hand. This talk explores how machine learning can help navigate this maze of data and turn raw measurements into chemical understanding.

Furthermore, drawing on work in high-resolution mass spectrometry, reaction discovery, molecular machine learning, and autonomous chemical research, a broader vision for the role of AI in chemistry will be described. Across these areas, the central idea is the same: learning algorithms can uncover structure in complex experimental data, connect measurements to molecular and mechanistic insight, and help scientists reason more effectively about chemical systems. Rather than treating machine learning as a separate computational layer, this work aims to integrate it directly into the process of chemical measurement, interpretation, and discovery.