And what I’m saying is that based on the code you provided, you should not expect it to work in general, because differentiating through noise is undefined behavior in the majority of cases
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Any faster way of computing small gradients? | 21 | 2119 | August 11, 2022 | |
| Speeding up gradient of logpdf | 19 | 870 | February 12, 2024 | |
| DiffEqGPU.jl with CUDA: Error computing gradients through SDE solver | 12 | 352 | January 10, 2024 | |
| Understanding and optimizing Enzyme.jl Reverse AD on CUDA | 5 | 203 | October 25, 2025 | |
| Correct sensealg for an ensemble problem of stochastic UODEs with diagonal noise | 17 | 1016 | November 23, 2020 |