Moment Estimate and Variational Approach for Learning Generalized Diffusion with Non-gradient Structures
By: Fanze Kong, Chen-Chih Lai, Yubin Lu
Potential Business Impact:
Finds hidden rules in how things move.
This paper proposes a data-driven learning framework for identifying governing laws of generalized diffusions with non-gradient components. By combining energy dissipation laws with a physically consistent penalty and first-moment evolution, we design a two-stage method to recover the pseudo-potential and rotation in the pointwise orthogonal decomposition of a class of non-gradient drifts in generalized diffusions. Our two-stage method is applied to complex generalized diffusion processes including dissipation-rotation dynamics, rough pseudo-potentials and noisy data. Representative numerical experiments demonstrate the effectiveness of our approach for learning physical laws in non-gradient generalized diffusions.
Similar Papers
Moment Estimates and DeepRitz Methods on Learning Diffusion Systems with Non-gradient Drifts
Machine Learning (CS)
Learns how things move and change, even with messy data.
Neural non-canonical Hamiltonian dynamics for long-time simulations
Machine Learning (CS)
Teaches computers to predict how things move.
Learning Generalized Hamiltonian Dynamics with Stability from Noisy Trajectory Data
Machine Learning (CS)
Teaches computers to predict how things move.