The implementation of a 3-pass method for fast automatic differentiation

Andrei Y. Gorchakov


In this paper, we consider the implementation of the three-pass method of fast automatic differentiation. Adding one more pass to the standard 2-pass method leads to a significant decrease in the consumption of RAM, with an insignificant (25%) increase in the number of calculations. Estimates of performance and requirements for the amount of RAM of 2 and 3-pass methods are given. A numerical experiment is performed on the example of the inverse problem of reconstructing the initial data for the transport equation.

Full Text:

PDF (Russian)


Ajda-Zade K.R., Evtushenko Ju.G. Bystroe avtomaticheskoe differencirovanie // Matematicheskoe modelirovanie, 1989. V. 1, pp. 121-139.

Evtushenko Ju.G., Mazurik V.P. Matematicheskoe obespechenie optimizacii. M: Znanie. 1989.

Evtushenko Y.G. Automatic differentiation viewed from optimal control theory. Proceedings of the Workshop on Automatic Differentiation of Algorithms: Theory, Implementation and Application. Philadelphia: SIAM, 1991.

Griewank A, Corliss G.F., eds. Automatic Differentiation of Algoritms. Theory, Implementation and Application. Philadelphia: SIAM, 1991.

Gorchakov A.Ju., Posypkin M.A. Jeffektivnost' metodov lokal'nogo poiska v zadache minimizacii jenergii ploskogo kristalla. Mezhdunarodnyj nauchnyj zhurnal «sovremennye informacionnye tehnologii i IT-obrazovanie». – 2017. – V. 13. – N. 2. – pp. 97-102.

Evtushenko, Ju. G., Lur'e, S. A., Posypkin, M. A., Soljaev, Ju. O. (2016). Primenenie metodov optimizacii dlja poiska ravnovesnyh sostojanij dvumernyh kristallov. Zhurnal vychislitel'noj matematiki i matematicheskoj fiziki, 56(12), 2032-2041.

Turkin A., Thu A. Benchmarking Python Tools for Automatic Differentiation. International Journal of Open Information Technologies. – 2016. – V. 4. – N. 9.

Hogan R. J. Fast reverse-mode automatic differentiation using expression templates in C++. ACM Transactions on Mathematical Software (TOMS). – 2014. – V. 40. – N. 4. – pp. 26.

Griewank A., Walther A. Algorithm 799: revolve: an implementation of checkpointing for the reverse or adjoint mode of computational differentiation. ACM Transactions on Mathematical Software (TOMS). 2000. V. 26. N. 1. pp. 19-45.

Evtushenko Ju.G. Optimizacija i bystroe avtomaticheskoe differencirovanie. Nauchnoe izdanie VC RAN, 2013.


  • There are currently no refbacks.

Abava   MSU conference 2018

ISSN: 2307-8162