PDE-free mass-constrained learning of complex systems with hidden states

Keyphrases

• Partial Differential Equations 100%
• Hidden States 100%
• Reduced Order Model 100%
• Equation-free 100%
• Constrained Learning 100%
• Proper Orthogonal Decomposition 75%
• Solution Operator 75%
• Diffusion Maps 50%
• High-dimensional Space 50%
• Spatiotemporal Evolution 50%
• Latent Space 50%
• Explicit Model 25%
• Spatiotemporal Dynamics 25%
• K-nearest Neighbor (K-NN) 25%
• Benchmark Problems 25%
• Stable Approximation 25%
• Walking Time 25%
• Multivariate Autoregressive Model 25%
• Velocity Field 25%
• Long Time Horizon 25%
• Nonlinear Manifold Learning 25%
• Approximation Accuracy 25%
• Interpolation Method 25%
• Approximate Solution 25%
• Crowd Dynamics 25%
• Machine Learning Framework 25%
• Latent Representation 25%
• Manifold Learning Algorithm 25%
• Highest Density Region 25%
• Preimage Problem 25%
• Navier-Stokes 25%
• Lifting Operator 25%
• Avoiding Obstacles 25%
• Unknown Mass 25%
• Sparse Identification of Nonlinear Dynamics (SINDy) 25%
• Obstacle Density 25%
• Convex Interpolation 25%
• KNN Algorithm 25%
• Hughes Model 25%
• Latent Dynamics 25%
• CFD Problems 25%
• Parsimonious Model 25%
• Stokes Velocity 25%
• Passive Tracer 25%

Mathematics

• Hidden State 100%
• Partial Differential Equation 100%
• Complex System 100%
• Manifold 50%
• Dimensional Space 50%
• Approximates 25%
• Minimizes 25%
• Autoregressive Model 25%
• Pre-Image 25%
• Velocity Distribution 25%
• Interpolation 25%
• Convex 25%