Provably correct, asymptotically efficient, higher-order reverse-mode automatic differentiation
View / Open Files
Publication Date
2022-01-16Journal Title
Proceedings of the ACM on Programming Languages
ISSN
2475-1421
Publisher
Association for Computing Machinery (ACM)
Volume
6
Issue
POPL
Pages
1-30
Type
Article
This Version
VoR
Metadata
Show full item recordCitation
Krawiec, F., Peyton Jones, S., Krishnaswami, N., Ellis, T., Eisenberg, R., & Fitzgibbon, A. (2022). Provably correct, asymptotically efficient, higher-order reverse-mode automatic differentiation. Proceedings of the ACM on Programming Languages, 6 (POPL), 1-30. https://doi.org/10.1145/3498710
Abstract
<jats:p>In this paper, we give a simple and efficient implementation of reverse-mode automatic differentiation, which both extends easily to higher-order functions, and has run time and memory consumption linear in the run time of the original program. In addition to a formal description of the translation, we also describe an implementation of this algorithm, and prove its correctness by means of a logical relations argument.</jats:p>
Identifiers
External DOI: https://doi.org/10.1145/3498710
This record's URL: https://www.repository.cam.ac.uk/handle/1810/334648
Rights
Attribution-NoDerivatives 4.0 International
Licence URL: https://creativecommons.org/licenses/by-nd/4.0/
Statistics
Total file downloads (since January 2020). For more information on metrics see the
IRUS guide.