DPM: A deep learning PDE augmentation method with application to large-eddy simulation

Mathematics

• Large Eddy Simulation 100%
• Augmentation 81%
• Training 72%
• Learning 62%
• Partial differential equation 49%
• Physics 47%
• Neural Networks 21%
• Filter 20%
• Direct numerical Simulation 16%
• Overfitting 15%
• Unknown 14%
• Leverage 14%
• Fidelity 14%
• Divergence-free 13%
• Loss Function 11%
• Experimental Data 11%
• Gradient 8%
• Optimization 8%
• Numerical Solution 8%
• Simulation 7%
• Design 7%
• Framework 6%
• Model 3%

Physics & Astronomy

• large eddy simulation 76%
• learning 74%
• education 67%
• partial differential equations 67%
• augmentation 41%
• physics 36%
• closures 22%
• filters 16%
• incompressibility 15%
• pulse detonation engines 14%
• machine learning 14%
• isotropic turbulence 14%
• direct numerical simulation 11%
• divergence 10%
• turbulence 8%
• formulations 7%
• optimization 7%
• gradients 7%
• simulation 4%

Engineering & Materials Science

• Large eddy simulation 76%
• Partial differential equations 67%
• Deep learning 61%
• Physics 59%
• Turbulence 24%
• Neural networks 16%
• Relaxation 12%
• Machine learning 10%