arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.12895math.NAcs.LGcs.NA
收藏

高维全非线性PIDE和耦合跳变FBSDES的深度学习数值方法

Deep learning numerical methods for high-dimensional fully nonlinear PIDEs and coupled FBSDEs with jumps

Shanghai Normal University(上海师范大学);Hainan University(海南大学)

Wansheng Wang, Jie Wang, Jinping Li, Feifei Gao, Yi Fu

摘要:

提出了一种求解高维抛物型积分微分方程(PIDE)和高维带跳跃的正倒向随机微分方程(FBSDEJ)的深度学习算法,其中跳跃-扩散过程由布朗运动和独立补偿泊松随机测度导出.在该算法中,一对深度神经网络的梯度和积分核的近似是在一个关键的方式引入基于深度FBNN方法。为了推导这种深度学习算法的误差估计,研究了马尔可夫迭代的收敛性、欧拉时间离散化的误差界以及深度学习算法的模拟误差。两个数值例子表明了该算法的有效性。

英文摘要:

We propose a deep learning algorithm for solving high-dimensional parabolic integro-differential equations (PIDEs) and high-dimensional forward-backward stochastic differential equations with jumps (FBSDEJs), where the jump-diffusion process are derived by a Brownian motion and an independent compensated Poisson random measure. In this novel algorithm, a pair of deep neural networks for the approximations of the gradient and the integral kernel is introduced in a crucial way based on deep FBSDE method. To derive the error estimates for this deep learning algorithm, the convergence of Markovian iteration, the error bound of Euler time discretization, and the simulation error of deep learning algorithm are investigated. Two numerical examples are provided to show the efficiency of this proposed algorithm.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑