Re: Differentiable Forth

AD is a powerful method when you need the derivatives of calculated
(not measured) functions. But if you don't know what a Jacobian
matrix is, you probably won't need AD.
There is a whole website dedicated to AD, tools and applications:
https://www.autodiff.org/
Projected to Forth, an autodiff wordset could be thought of similar
to a complex number wordset imaging the standard fp-number wordset.
However, care must be taken with non-differentiable regions in
e.g. fabs, fsqrt and some (inverse) trigonometric functions.
(AD libraries in C++ use operator overloading)