How to Train Your Neural ODE: the World of Jacobian and Kinetic Regularization.
Chris FinlayJörn-Henrik JacobsenLevon NurbekyanAdam M. ObermanPublished in: ICML (2020)
Keyphrases
- network architecture
- neural network
- neural model
- vector field
- bio inspired
- parameter selection
- monte carlo
- singular value decomposition
- biologically inspired
- ordinary differential equations
- nonlinear predictive control
- neural computation
- neural fuzzy
- regularization method
- world model
- data dependent
- regularization parameter
- experimental data
- semi supervised learning