Neural Ordinary Differential Equations
From MaRDI portal
Abstract: We introduce a new family of deep neural network models. Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a black-box differential equation solver. These continuous-depth models have constant memory cost, adapt their evaluation strategy to each input, and can explicitly trade numerical precision for speed. We demonstrate these properties in continuous-depth residual networks and continuous-time latent variable models. We also construct continuous normalizing flows, a generative model that can train by maximum likelihood, without partitioning or ordering the data dimensions. For training, we show how to scalably backpropagate through any ODE solver, without access to its internal operations. This allows end-to-end training of ODEs within larger models.
Cited in
(20)- tfNeuralODE
- Learning dynamical systems from noisy data with Weak-form Kernel Ridge Regression
- Convolutional Symmetric AutoEncoders: enhancing latent stability via differential geometry
- Data-Adaptive Learning of Dynamical Systems by Matching Transfer Operators and Invariant Measures
- An optimal control approach for neural network architecture adaptation with a posteriori error estimation
- A unified perspective of Gaussian process approximation for differential equations
- CSympNet-ID: conformal-symplectic map learning for linearly damped Hamiltonian systems
- Positive-Allocation Companion Predictors for Nonlinear Dynamics and Their Finite-Difference Diagnostics
- Lightweight return-mapping surrogates for multiscale plasticity: a practical guide
- Perturbative-NeuSA: A Structured Spectral Framework for Time-Dependent PDEs
- A neural operator view on U-Nets for inverse imaging problems
- Physics-informed reduced-order modelling with equivariant spectral submanifolds
- Differentiate the Solver, Not the Equation: Reverse-Sweep Adjoints for Block Implicit Simulation
- Kernel Localization and Whole-Trajectory Generalization for Linear Multistep Methods in Deep Learning-Based Discovery of Dynamical Systems
- Learning Informative Prior with Infinite-Dimensional Continuous Normalizing Flow for Bayesian Inverse Problem
- Physics-Guided Conditional Flow Matching with Energy Regularization for Robust PDE Inverse Problems
- Deep operator learning for efficient sampling from invariant measures of stochastic differential equations
- Integral chemical reaction neural networks
- NEXT: Physics-Informed Neuro-Spectral Exponential Time Differencing Architectures
- Hybrid Joint-Selective Optimization: Reduced-Space Levenberg-Marquardt Refinement of Low-Dimensional Parameters of Interest
This page was built for publication: Neural Ordinary Differential Equations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5980956)
