JuliaDiff / JuliaDiff/ForwardDiff.jl
Efficient partial Hessian calculation
Nobody has claimed this yet.
- Dominant language
- Julia
- Stars
- 1k
- Forks
- 160
- PR merge metrics
- No merged PRs in 30d
Description
Suppose I have a function F(x,y) of two variables x and y where dim(x) << dim(y). What is the most efficient way to calculate both ForwardDiff.jacobian(w1 -> ForwardDiff.gradient(w2 -> F(w2, y), w1), x) and
ForwardDiff.jacobian(z -> ForwardDiff.gradient(w -> F(w, z), x),y)? These work, but I'm wondering if there's something faster. I think ForwardDiff.hessian would be way to expensive since I have no need for the second derivative w.r.t. y. I just need the partials \nabla_xxF and \nabla_xyF. Thanks!
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No source file or test is identified. Start by reviewing the Julia ForwardDiff.jacobian, gradient, and hessian APIs used in the report, then determine whether the requested partial Hessian calculation fits an existing entry point. Done means an agreed efficient approach is implemented or documented and its behavior is covered by relevant tests.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- julia
- Domain
- devtools
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100