So we started a new huge project for which we aim to use Optimization.jl with Enzyme.jl as AD backend. We came up with some early prototypes and plugged that into Optimization.jl, unfortunately not very succesful (It seems to be the case that the closures we need to provide to Optimization.jl intro…

For arrays in reverse mode, you should not use Active(x), but rather create a zero shadow of memory for it to store the derivatives into. In other words, something like: dx = zeros(4) autodiff(Reverse,vec_input_target,Active,Duplicated(x, dx)) # result is in dx

Autodiff of vector inputs with Enzyme.jl (and possibly Optimization.jl)

Specific Domains Optimization (Mathematical)

wsmoses August 15, 2023, 7:39pm 8

That’s the error message as here, so I’ll link my answer here too: Understanding an Enzyme Warning Message - #2 by wsmoses

Topic		Replies	Views
Reliability of Enzyme.jl General Usage question , diffeq , autodiff	11	2018	October 23, 2022
Autodiff for max-flow Modelling & Simulations autodiff , graphs , enzyme	14	320	May 18, 2024
End-to-end gradients with Enzyme when training ODE parameters General Usage diffeq , autodiff , enzyme	1	322	October 21, 2022
How to differentiate using Enzyme.jl? General Usage autodiff , enzyme	7	180	November 18, 2024
Using NonlinearSolve with autodiff Optimization (Mathematical) zygote , autodiff , enzyme , nonlinearsolve	9	207	October 17, 2024

Autodiff of vector inputs with Enzyme.jl (and possibly Optimization.jl)

Related topics