Multitask Deep Ensemble Prediction of Molecular Energetics in Solution: From Quantum Mechanics to Experimental Properties

Keyphrases

• Quantum Mechanics 100%
• Multi-task 100%
• Ensemble Forecast 100%
• Electronic Energy 100%
• Experimental Properties 100%
• Molecular Energetics 100%
• Deep Ensemble 100%
• Atomic Contributions 66%
• Prediction Error 33%
• Experimental Dataset 33%
• Chemical Environment 33%
• Model Uncertainty 33%
• Force Field 33%
• Art Performance 33%
• Geometry Optimization 33%
• Molecular Conformation 33%
• Heavy Atoms 33%
• Machine Learning Techniques 33%
• Density Functional Theory 33%
• Transfer Free Energy 33%
• Quantum Mechanical Methods 33%
• Solvation Free Energy 33%
• Prediction Task 33%
• B3LYP/6-31G 33%
• Continuum Solvent Model 33%
• Machine Learning Models 33%
• Gas-water 33%
• Deep Learning Model 33%
• Octanol 33%
• Chemical Accuracy 33%
• Hydration Free Energy 33%
• Chemical Knowledge 33%

Chemistry

• Energetics 100%
• Statistical Ensemble 100%
• Gibbs Free Energy 100%
• Quantum Mechanics 100%
• Logp 66%
• Solvation 33%
• Density Functional Theory 33%
• DFT-B3LYP Calculation 33%
• Quantum Mechanical Method 33%
• Electron Energy Level 33%
• Octanol 33%