Understanding the Difficulty of Training Physics-Informed Neural Networks on Dynamical Systems.
Franz M. RohrhoferStefan PoschClemens GößnitzerBernhard C. GeigerPublished in: CoRR (2022)
Keyphrases
- dynamical systems
- neural network
- training process
- training algorithm
- differential equations
- dynamic systems
- nonlinear dynamical systems
- linear systems
- multi layer perceptron
- nonlinear dynamics
- phase space
- state space
- partially observable
- control theory
- agent environment
- qualitative simulation
- predictive state representations
- dynamical behavior
- dynamical models
- training set
- partially observable markov decision processes
- qualitative reasoning
- test set
- semi supervised
- reinforcement learning