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

by Côme Huré, Huyên Pham, Achref Bachouch & Nicolas Langrené

This paper presents several numerical applications of deep learning-based algorithms that have been analyzed in [Bachouch et al., 2018a]. 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 [Weinan et al. 2017] and on quadratic Backward Stochastic Differential equations as in [Chassagneux et al., 2016]. We also provide numerical results for an option hedging problem in finance, and energy storage problems arising in the valuation of gas storage and in microgrid management.

Download the paper

More working papers

This entry was posted in Working Papers. Bookmark the permalink.