Neural network ensemble for computing cross sections for rotational transitions in H${}_2$O + H${}_2$O collisions

Abstract

Water (H${}_2$O) is one of the most abundant molecules in the universe and is found in a wide variety of astrophysical environments. Rotational transitions in H${}_2$O + H${}_2$O collisions are important in modeling environments rich in water molecules but they are computationally intractable using quantum mechanical methods. Here, we present a machine learning (ML) tool using an ensemble of neural networks (NNs) to predict cross sections to construct a database of rate coefficients for rotationally inelastic transitions in collisions of complex molecules such as water. The proposed methodology utilizes data computed with a mixed quantum-classical theory (MQCT). We illustrate that efficient ML models using NN can be built to accurately interpolate in the space of 12 quantum numbers for rotational transitions in two asymmetric top molecules, spanning both initial and final states. We examine various architectures of data corresponding to each collision energy, symmetry of water molecule, and excitation/de-excitation rotational transitions, and optimize the training/validation data sets. Using only about 10\% of the computed data for training, the NNs predict cross sections of state-to-state rotational transitions of H${}_2$O + H${}_2$O collision with average relative root mean square error of 0.409. Thermally averaged cross sections, computed using the predicted state-to-state cross sections ($\sim$90\%) and the data used for training and validation ($\sim$10\%) were compared against those obtained entirely from MQCT calculations. The agreement is found to be excellent with an average percent deviation of about $\sim$13.5\%. The methodology is robust, and thus, applicable to other complex molecular systems.

13 pages, 10 figures, journal article, submitted to PCCP

References in corpus (31)

• A computer program for fast non-LTE analysis of interstellar line spectra
• An introduction to quantum machine learning
• Challenges and Opportunities in Quantum Machine Learning
• Quantum machine learning: a classical perspective
• Molecular excitation in the Interstellar Medium: recent advances in collisional, radiative and chemical processes
• Water in Star-Forming Regions with the Herschel Space Observatory (WISH): Overview of key program and first results
• LIME - a flexible, non-LTE line excitation and radiation transfer method for millimeter and far-infrared wavelengths
• Water in star-forming regions (WISH): Physics and chemistry from clouds to disks as probed by Herschel spectroscopy
• MOLSCAT: a program for non-reactive quantum scattering calculations on atomic and molecular collisions
• Machine Learning and LHC Event Generation
• Bayesian machine learning for quantum molecular dynamics
• Transformer Quantum State: A Multi-Purpose Model for Quantum Many-Body Problems
• Machine learning discovery of new phases in programmable quantum simulator snapshots
• Gaussian process model of 51-dimensional potential energy surface for protonated imidazole dimer
• Automatic Classification of Galaxy Morphology: a rotationally invariant supervised machine learning method based on the UML-dataset
• A unique, ring-like radio source with quadrilateral structure detected with machine learning
• Machine-learning-corrected quantum dynamics calculations
• Machine Learning methods to estimate observational properties of galaxy clusters in large volume cosmological N-body simulations
• Neural Networks Optimized by Genetic Algorithms in Cosmology
• Machine Learning for Observables: Reactant to Product State Distributions for Atom-Diatom Collisions
• MOLPOP-CEP: An Exact, Fast Code for Multi-Level Systems
• A flexible event reconstruction based on machine learning and likelihood principles
• Deep learning predicted elliptic flow of identified particles in heavy-ion collisions at the RHIC and LHC energies
• Submillimeter and Millimeter Masers
• A novel Machine-Learning method for spin classification of neutron resonances
• Improved temperature dependence of rate coefficients for rotational state-to-state transitions in H${}_2$O + H${}_2$O collisions
• No Headache for PIPs: A PIP Potential for Aspirin Outperforms Other Machine-Learned Potentials
• Quantum Gaussian process model of potential energy surface for a polyatomic molecule
• Mixed Quantum/Classical Theory for Rotational Energy Exchange in Symmetric-Top-Rotor + Linear-Rotor Collisions and a Case Study of $\mathrm{N}{\mathrm{D}}_3+{\mathrm{D}}_2$ System
• Rate Coefficients for Rotational State-to-State Transitions in H${}_2$O + H${}_2$ Collisions as Predicted by Mixed Quantum/Classical Theory (MQCT)
• On Mixed Quantum/Classical Theory for Rotationally Inelastic Scattering of Identical Collision Partners