Neural Petri flows for chemical reactions
Presents a mathematically rigorous ML approach for computational chemistry with proven performance advantages over existing methods.
AI Summary
Researchers introduce Neural Petri Flow, a novel architecture that preserves Petri net semantics while learning chemical reaction rates, achieving state-of-the-art performance on atom mapping and reaction prediction tasks.
Excerpt
Petri nets have been used to describe chemical processes such as reactions.They map well to chemistry: Places are the bonds between atoms and the free valence of each atom, a token is a unit of bond order, a transition forms or breaks a bond, the conserved quantities are the valence budgets of the atoms, and the enabling rule is the valence rule. These semantics are not guaranteed by learned models of reactions or neural networks that are built on Petri nets that use the net as a scaffold for me