PEACE: Covariant learning of nonadiabatic manifolds with parity-resolved Hamiltonians

Key points

• Here we introduce PEACE, which combines a parity-equivariant latent Hamiltonian with a learned electronic connection.1
• Controlled ablations reveal the complementary roles of symmetry-allowed state mixing and electronic-frame variation in reproducing crossing structures and relaxation dynamics.1
• PEACE closely reproduces excited-state population dynamics from first-principles simulations, while its extension to spin-orbit coupling enables simulations of intersystem crossing.1
• These results demonstrate that a more complete incorporation of the underlying physics into learned electronic representations leads to more accurate predictions of nonadiabatic dynamics.1

Sources (1)

• [1]PEACE: Covariant learning of nonadiabatic manifolds with parity-resolved Hamiltonians
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 07:21 AM

  Nonadiabatic molecular dynamics provides mechanistic insight into light-driven processes and informs the design of molecules and materials for solar energy conversion, photocatalysis and photo switching.

  Here we introduce PEACE, which combines a parity-equivariant latent Hamiltonian with a learned electronic connection.

Extractive summary: sentences quoted from the sources.