Using an Autoencoder for Dimensionality Reduction in Quantum Dynamics [chapter]

Sebastian Reiter, Thomas Schnappinger, Regina de Vivie-Riedle
2019 Lecture Notes in Computer Science  
A key step in performing quantum dynamics for a chemical system is the reduction of dimensionality to allow a numerical treatment. Here, we introduce a machine learning approach for the (semi)automatic construction of reactive coordinates. After generating a meaningful data set from trajectory calculations, we train an autoencoder to find a lowdimensional set of non-linear coordinates for use in molecular quantum dynamics. We compare the wave packet dynamics of proton transfer reactions in both
more » ... linear and non-linear coordinate spaces and find significant improvement for physical properties like reaction timescales.
doi:10.1007/978-3-030-30493-5_73 fatcat:2lqnldg4tfe4nglk7qp7ir6d7m