An acausal modeling framework for automatically parallelized scientific machine learning (SciML) in Julia. A computer algebra system for integrated symbolics for physics-informed machine learning and automated transformations of differential equations
This can really help enforce conditions like non-negativity since if your space doesn't allow it, then the numerical solver can guarantee it won't happen if it solves for exp(x) instead of x.
Basic things like http://www.math.mcgill.ca/jakobson/courses/ma261/lecture5.pdf
This can really help enforce conditions like non-negativity since if your space doesn't allow it, then the numerical solver can guarantee it won't happen if it solves for
exp(x)
instead ofx
.