A deep learning-based Monte Carlo algorithm with applications in American options pricing

This is the authors' abstract. We don't add key points for this paper.

.In this article, we introduce a deep learning-based Monte Carlo method and apply it to pricing high-dimensional American options. This method keeps track of the approximating optimal stopping time and evaluates it by independent samples at each time step. The consistency and convergence rate of the algorithm are derived, which show that this method is able to circumvent the curse of dimensionality. The numerical results show that our method is efficient and accurate in pricing high-dimensional American options.