Uncertainty quantification in molecular simulations with dropout neural network potentials

Mathematics

• Molecular Simulation 100%
• Uncertainty Quantification 78%
• Drop out 77%
• Neural Networks 55%
• Uncertainty 49%
• Interatomic Potential 30%
• Density Functional Theory 24%
• Machine Learning 18%
• Training 17%
• Graphene 12%
• Phonon 12%
• First-principles 10%
• Bayesian Statistics 10%
• Carbon 10%
• Computational Cost 8%
• Predict 7%
• Predictors 7%
• Estimate 7%
• Configuration 6%
• Statistics 5%
• Energy 5%
• Unknown 5%
• Design 5%
• Demonstrate 5%
• Class 2%

Chemical Compounds

• Interatomic Potential 86%
• Models (Physical) 39%
• Simulation 37%
• Density Functional Theory 36%
• Phonon 32%
• Potential Energy 29%
• Graphene 20%
• Force 19%
• Carbon Atom 13%

Engineering & Materials Science

• Uncertainty 36%
• Neural networks 35%
• Density functional theory 24%
• Machine learning 12%
• Graphene 9%
• Potential energy 9%
• Carbon 6%
• Statistics 6%
• Costs 3%