Learning reduced kinetic Monte Carlo models of complex chemistry from molecular dynamics

  title={Learning reduced kinetic Monte Carlo models of complex chemistry from molecular dynamics},
  author={Qian Yang and Carlos A. Sing-Long and Evan J Reed},
  journal={Chemical Science},
  pages={5781 - 5796}
We propose a novel statistical learning framework for automatically and efficiently building reduced kinetic Monte Carlo (KMC) models of large-scale elementary reaction networks from data generated by a single or few molecular dynamics simulations (MD). 
Data-Driven Methods for Building Reduced Kinetic Monte Carlo Models of Complex Chemistry from Molecular Dynamics Simulations
Complex chemical processes such as those found in combustion, the decomposition of energetic materials, and the chemistry of planetary interiors, are typically studied at the atomistic level using
Atomic-Level Features for Kinetic Monte Carlo Models of Complex Chemistry from Molecular Dynamics Simulations.
Using the complex chemical network of hydrocarbon pyrolysis as an example, it is demonstrated that kinetic models built using atomic features are able to explore chemical reaction pathways never observed in the MD simulations used to parameterize them, a critical feature to describe rare events.
Atomic-level features for statistical learning of the chemical kinetics from molecular dynamics simulations
Atomic-level features are shown to construct reaction mechanisms and estimate reaction rates of unknown molecular species from elementary atomic events and it is found that atomic features can better describe the formation of large molecules enabling the simultaneous description of small molecules and condensed phases.
Graph-Convolutional Neural Networks for Hydrocarbons Kinetics Predictions
The kinetics of the pyrolysis of hydrocarbons is predicted using the powerful framework of Graph Convolutional Neural Networks (GCNN), which is particularly well suited for describing molecules and crystals.
Rapid data-driven model reduction of nonlinear dynamical systems including chemical reaction networks using ℓ1-regularization.
A new data-driven method utilizing ℓ1-regularization for model reduction of nonlinear dynamical systems, which involves minimal parameterization and has polynomial-time complexity, allowing it to easily handle large-scale systems with as many as thousands of components in a matter of minutes.
Heuristics for chemical species identification in dense systems.
A new approach to identify chemical species from molecular dynamics (MD) simulations of reacting materials under extreme temperatures and pressures is presented. The approach is based on
Predicting Molecule Size Distribution in Hydrocarbon Pyrolysis using Random Graph Theory
Hydrocarbon pyrolysis is a complex process involving large numbers of chemical species and types of chemical reactions. Its quantitative description is important for planetary sciences, in
Autonomous reinforcement learning agent for chemical vapor deposition synthesis of quantum materials
Predictive materials synthesis is the primary bottleneck in realizing functional and quantum materials. Strategies for synthesis of promising materials are currently identified by time-consuming
Physics-Constrained, Data-Driven Discovery of Coarse-Grained Dynamics
This paper advocates the paradigm of data-driven discovery for extract- ing governing equations by employing fine-scale simulation data and advocates a sparse Bayesian learning perspective which avoids overfitting and reveals the most salient features in the CG evolution law.
Quantum Material Synthesis by Reinforcement Learning
Designing conditions for experimental synthesis is the primary bottleneck for the realization of new functional quantum materials. Current strategies to synthesize new promising materials with


Kinetic mechanism reduction based on an integer programming approach
An approach toward kinetic mechanism reduction, both in terms of reactions and species, is discussed and several examples, utilizing a variety of kinetic networks, are presented and the results are analyzed.
Efficient maximum likelihood parameterization of continuous-time Markov processes.
A maximum likelihood estimator is introduced for constructing continuous-time Markov processes over finite state-spaces from data observed at a finite time interval that is dramatically more efficient than prior approaches, enables the calculation of deterministic confidence intervals in all model parameters, and can easily enforce important physical constraints on the models.
Reduction of dynamical biochemical reactions networks in computational biology
The application of model reduction to the problem of parameter identification, via backward pruning machine learning techniques is discussed, and the main approaches to model reduction of reaction networks are revisited, including quasi-steady state (QSS) and quasi-equilibrium approximations (QE).
Modeling and Simulating Chemical Reactions
Some of the basic concepts of deterministic reaction rate equations are introduced in an accessible manner and some challenges that currently occupy researchers in this area are pointed to.
Model reduction for chemical kinetics: an optimization approach
An optimization-based method for reduction of the number of species and reactions in chemical kinetics models is described and numerical results for several reaction mechanisms illustrate the potential of this approach.
Neural Networks for the Prediction of Organic Chemistry Reactions
This work explores the use of neural networks for predicting reaction types, using a new reaction fingerprinting method and combines this predictor with SMARTS transformations to build a system which, given a set of reagents and reactants, predicts the likely products.
Discovering chemistry with an ab initio nanoreactor
Using the nanoreactor, new pathways for glycine synthesis from primitive compounds proposed to exist on the early Earth are shown, providing new insight into the classic Urey-Miller experiment, highlighting the emergence of theoretical and computational chemistry as a tool for discovery in addition to its traditional role of interpreting experimental findings.
Exact Stochastic Simulation of Coupled Chemical Reactions
There are two formalisms for mathematically describing the time behavior of a spatially homogeneous chemical system: The deterministic approach regards the time evolution as a continuous, wholly