Deep neural networks algorithms for stochastic control problems on finite horizon: numerical applications - Archive ouverte HAL Access content directly
Journal Articles Methodology and Computing in Applied Probability Year : 2020

Deep neural networks algorithms for stochastic control problems on finite horizon: numerical applications

(1) , (2) , (3) , (2)
1
2
3

Abstract

This paper presents several numerical applications of deep learning-based algorithms that have been introduced in [HPBL18]. Numerical and comparative tests using TensorFlow illustrate the performance of our different algorithms, namely control learning by performance iteration (algorithms NNcontPI and ClassifPI), control learning by hybrid iteration (algorithms Hybrid-Now and Hybrid-LaterQ), on the 100-dimensional nonlinear PDEs examples from [EHJ17] and on quadratic backward stochastic differential equations as in [CR16]. We also performed tests on low-dimension control problems such as an option hedging problem in finance, as well as energy storage problems arising in the valuation of gas storage and in microgrid management. Numerical results and comparisons to quantization-type algorithms Qknn, as an efficient algorithm to numerically solve low-dimensional control problems, are also provided; and some corresponding codes are available on https://github.com/comeh/.
Fichier principal
Vignette du fichier
Deepconsto-Partie2_rev4.pdf (3.46 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01949221 , version 1 (12-12-2018)
hal-01949221 , version 2 (17-05-2019)
hal-01949221 , version 3 (25-01-2020)

Identifiers

Cite

Achref Bachouch, Côme Huré, Nicolas Langrené, Huyen Pham. Deep neural networks algorithms for stochastic control problems on finite horizon: numerical applications. Methodology and Computing in Applied Probability, In press. ⟨hal-01949221v3⟩
474 View
1134 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More