Differentiable Programming for Differential Equations: A Review
math.NA, cs.NA, math.DS, physics.comp-ph, stat.ML
Submitted: 2024-06-14
Updated: 2025-10-29
Code: https://github.com/ODINN-SciML/DiffEqSensitivity
Terminology
Sources
- Differentiable quantum computational chemistry with PennyLane
- Julia: A Fast Dynamic Language for Technical Computing
- On the complexity of nonsmooth automatic differentiation
- Toward the End-to-End Optimization of Particle Physics Instruments with Differentiable Programming: a White Paper
- A Differentiable Programming System to Bridge Machine Learning and Scientific Computing
- On the Equivalence of Automatic and Symbolic Differentiation
- Efficient Automatic Differentiation of Implicit Functions
- Review of theory and implementation of hyper-dual numbers for first and second order automatic differentiation
- Fast differentiation of hyperbolic chaos
- Faster Training of Neural ODEs Using Gau{\ss}-Legendre Quadrature
- Discretize-Optimize vs. Optimize-Discretize for Time-Series Regression and Continuous Normalizing Flows
- NonlinearSolve.jl: High-Performance and Robust Solvers for Systems of Nonlinear Equations in Julia
- Universal Differential Equations for Scientific Machine Learning
- Differentiable Physics: A Position Piece
- Forward-Mode Automatic Differentiation in Julia
- AbstractDifferentiation.jl: Backend-Agnostic Differentiable Programming in Julia
- Neural Stochastic Differential Equations: Deep Latent Gaussian Models in the Diffusion Limit
- NeuralPDE: Automating Physics-Informed Neural Networks (PINNs) with Error Approximations
Related papers
- Do physics-informed neural networks (PINNs) need to be deep? Shallow PINNs using the Levenberg-Marquardt algorithm
- A Neural-preconditioned Poisson Solver for Mixed Dirichlet and Neumann Boundary Conditions
- Advanced Linear Algebra with Applications - Part I (Numerical linear algebra for PDEs, machine learning, and data assimilation)
- Operator learning meets inverse problems: A probabilistic perspective
- Second-order consistency for learning chaotic dynamics via randomized Jacobian matching
- Windowed thinning and query complexity for the bouncy particle and Zigzag samplers