Computing the frequency fluctuation dynamics of highly coupled vibrational transitions using neural networks.

摘要: The description of frequency fluctuations for highly coupled vibrational transitions has been a challenging problem in physical chemistry. In particular, the complexity their Hamiltonian does not allow us to directly derive time evolution frequencies these systems. this paper, we present new approach by exploiting artificial neural network describe without relying on deconstruction Hamiltonian. To end, first explored use methodology predict amide I mode N-methylacetamide water. results show good performance compared with previous experimental and theoretical results. second part, is used investigate carbonyl stretch modes organic carbonates solvation shell lithium ion. case, fluctuation predicted networks shows agreement results, which suggests that model can be dynamics transitions.