Neural ODEs & Continuous-Time Models Reading List
Curated by Mouhssine Rifaki | Stanford Electrical Engineering | Last updated August 2026
Neural ODEs rethink depth as time, turning a ResNet into an ODE solved by adaptive integrators. The selection below covers the original formulation, its stable, second-order and jump extensions, and neighboring continuous-time models such as controlled differential equations and ODE processes.
Neural ODEs & Continuous-Time Models: 10 key papers
- Neural Ordinary Differential Equations
Chen et al. arXiv 2018.
- Augmented Neural ODEs
Dupont et al. arXiv 2019.
- FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Models
Grathwohl et al. arXiv 2018.
- Latent ODEs for Irregularly-Sampled Time Series
Rubanova et al. arXiv 2019.
- Neural Controlled Differential Equations for Irregular Time Series
Kidger et al. arXiv 2020.
- Dissecting Neural ODEs
Massaroli et al. arXiv 2020.
- Neural SDEs as Infinite-Dimensional GANs
Kidger et al. arXiv 2021.
- How to train your neural ODE: the world of Jacobian and kinetic regularization
Finlay et al. arXiv 2020.
- Hamiltonian Neural Networks
Greydanus et al. arXiv 2019.
- Lagrangian Neural Networks
Cranmer et al. arXiv 2020.
← Back to main page