pLaSDI learns a reduced governing equation for time-dependent NLTE atomic kinetics with physics-informed losses enforcing consistency, stability, and steady-state convergence, achieving <2% error on tin charge-state evolution at 5e4-1e5 speedup and stable extrapolation.
A comprehensive review of latent space dynamics identification algorithms for intrusive and non-intrusive reduced-order-modeling
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
roles
method 1polarities
use method 1representative citing papers
WGFINNs use weak-form loss functions with GENERIC structure preservation to recover governing equations more accurately from noisy observations than prior strong-form GFINNs.
Physics-conforming Latent Twins learns encoder-decoder pairs and latent flow maps that satisfy physical principles by design via constraint transfer and algebraic conditions on invariants and dissipation.
DIANO builds coarse-grid latent spaces for fluid dynamics data via neural operator encoding and decoding while integrating a differentiable PDE solver directly in the latent space for end-to-end physics-constrained training.
mLaSDI uses multi-stage residual decoder training with periodic activations to recover high-frequency details in latent space dynamics identification, yielding lower reconstruction and prediction errors than standard LaSDI for PDEs.
Higher-order LaSDI uses a high-order finite-difference scheme and rollout loss to improve long-term prediction accuracy in reduced-order models for parameterized PDEs, shown on the 2D Burgers equation.
MPEX AI Digital Twins is a project vision to train AI models on experimental and physics simulation data to create digital twins for material assessment metrics in plasma experiments.
citing papers explorer
-
Physics-Informed Latent Space Dynamics Identification for Time-Dependent NLTE Atomic Kinetics
pLaSDI learns a reduced governing equation for time-dependent NLTE atomic kinetics with physics-informed losses enforcing consistency, stability, and steady-state convergence, achieving <2% error on tin charge-state evolution at 5e4-1e5 speedup and stable extrapolation.
-
WGFINNs: Weak formulation-based GENERIC formalism informed neural networks
WGFINNs use weak-form loss functions with GENERIC structure preservation to recover governing equations more accurately from noisy observations than prior strong-form GFINNs.
-
Physics-conforming Latent Twins
Physics-conforming Latent Twins learns encoder-decoder pairs and latent flow maps that satisfy physical principles by design via constraint transfer and algebraic conditions on invariants and dissipation.
-
Differentiable Autoencoding Neural Operator for Interpretable and Integrable Latent Space Modeling
DIANO builds coarse-grid latent spaces for fluid dynamics data via neural operator encoding and decoding while integrating a differentiable PDE solver directly in the latent space for end-to-end physics-constrained training.
-
mLaSDI: Multi-stage latent space dynamics identification
mLaSDI uses multi-stage residual decoder training with periodic activations to recover high-frequency details in latent space dynamics identification, yielding lower reconstruction and prediction errors than standard LaSDI for PDEs.
-
Higher-Order LaSDI: Reduced Order Modeling with Multiple Time Derivatives
Higher-order LaSDI uses a high-order finite-difference scheme and rollout loss to improve long-term prediction accuracy in reduced-order models for parameterized PDEs, shown on the 2D Burgers equation.
-
MPEX AI Digital Twins
MPEX AI Digital Twins is a project vision to train AI models on experimental and physics simulation data to create digital twins for material assessment metrics in plasma experiments.