hackage.haskell.org : ad-delcont
Reverse-mode automatic differentiation using delimited continuations (shift/reset). The package exposes a small and easily extensible user interface to automatic differentiation combinators. It's also lightweight as a dependency, since it only requires base and transformers. To use the library, import Numeric.AD.DelCont, which also contains all documentation. Blog post : http://ocramz.github.io/haskell/automatic-differentiation/2021/07/19/ad-delcont.html References : F. Wang et al, Backpropagation with Continuation Callbacks : Foundations for Efficient and Expressive Differentiable Programming, NeurIPS 2018 - https://papers.nips.cc/paper/2018/file/34e157766f31db3d2099831d348a7933-Paper.pdf F. Wang et al, Demystifying Differentiable Programming : Shift/Reset the Penultimate Backpropagator, ICFP 2019 - https://www.cs.purdue.edu/homes/rompf/papers/wang-icfp19.pdf
Registry
-
Source
- JSON
purl: pkg:hackage/ad-delcont
Keywords:
bsd3
, library
, machine-learning
, math
, numeric
, optimisation
, optimization
, Propose Tags
, adjoint
, automatic-differentiation
, backpropagation
, gradient
License: BSD-3-Clause
Latest release: about 2 years ago
First release: about 4 years ago
Dependent repositories: 7
Downloads: 943 total
Stars: 15 on GitHub
Forks: 2 on GitHub
Total Commits: 51
Committers: 2
Average commits per author: 25.5
Development Distribution Score (DDS): 0.039
More commit stats: commits.ecosyste.ms
See more repository details: repos.ecosyste.ms
Funding links: https://github.com/sponsors/ocramz
Last synced: 27 days ago