Get Started
Research questionHow can GPU ODE ensemble solvers provide fast reverse-mode gradients without sacrificing solve speed?GPU ODE software often makes fast ensemble execution and efficient reverse-mode differentiation separate capabilities. This slows applications that repeatedly need parameter gradients while solving many independent trajectories.
Research Paper
Technology
Latest papersRecent research connected to this question, newest first.GRADSOLVE: fast exact gradients for ODE ensembles on GPUsThe evidence concerns explicit and Rosenbrock integrators in GRADSOLVE, with gradients computed by replaying recorded solver steps. Reported speedups over Diffrax vary by GPU generation, ensemble size, system stiffness, and target accuracy; the advantage narrows for large ensembles and stiff systems at tight accuracy.research paper · Sep 2, 2026
Related questions
How can we simulate and backpropagate through long coupled dynamical systems without stepwise time recursion?How can DeepONet-style neural operators represent sharp moving discontinuities in shock-dominated, low-viscosity PDEs?How can parametric PDE solution operators extrapolate to unseen regimes without failing silently?How can LLM-generated GPU kernels remain performant across hardware, inputs, and programming models?
Home
Topics
Search
Library