Department Seminar Series - Dr. Tyler Josephson

ATOMS: AI & Theory-Oriented Molecular Science

Computers have long assisted chemical, biochemical, and environmental engineers in solving problems. Recently, machine learning (ML) and artificial intelligence (AI) tools have exploded in performance and popularity, as machine learning algorithms can extract patterns from large datasets to generate predictions. In the AI & Theory-Oriented Molecular Science (ATOMS) Lab at UMBC, we're using ML algorithms and developing new ones, as we explore ways to develop artificial assistants to aid humans in scientific research.

Some ML tools lead to interpretable models, and some learning algorithms are informed by physics, but generally, these are not easily related to the equation-based theories and derivations in the literature. The development of artificial systems that operate on theories could open new avenues for scientific discovery, and we posit that such systems will need to have robust logical reasoning capabilities.

Automated reasoning tools can treat the “stuff” of theory as model inputs and outputs, perform logical transformations from one formula to another, and provide rigorous certifications when derivations are mathematically sound. We show how automated reasoning can test whether models generated from data are derivable from or otherwise related to some background theory. We also illustrate how recent interactive theorem proving tools are mature enough to formally prove complex mathematical statements, and we highlight the potential for these tools for not only formally checking scientific theories in our literature, but also for creating bug-free code for scientific computing.