TY - JOUR
T1 - Milstein-driven neural stochastic differential equation model with uncertainty estimates
AU - Zhang, Xiao
AU - Wei, Wei
AU - Zhang, Zhen
AU - Zhang, Lei
AU - Li, Wei
N1 - Publisher Copyright:
© 2023
PY - 2023/10
Y1 - 2023/10
N2 - Incorporating uncertainty quantification into the modeling of deep learning-based model has become a research focus in the deep learning community. Within this group of methods, stochastic differential equation (SDE)-based models have demonstrated advantages in their ability to model uncertainty quantification. However, the use of Euler's method in these models introduces imprecise numerical solutions, which limits the accuracy of SDE systems and weakens the performance of the network. In this study, we build a more precise Milstein-driven SDE network (MDSDE-Net) to improve the network performance. In addition, we analyze the convergence of the Milstein scheme and theoretically guarantee the feasibility of MDSDE-Net. Experimental and theoretical results show that the MDSDE-Net outperforms existing models.
AB - Incorporating uncertainty quantification into the modeling of deep learning-based model has become a research focus in the deep learning community. Within this group of methods, stochastic differential equation (SDE)-based models have demonstrated advantages in their ability to model uncertainty quantification. However, the use of Euler's method in these models introduces imprecise numerical solutions, which limits the accuracy of SDE systems and weakens the performance of the network. In this study, we build a more precise Milstein-driven SDE network (MDSDE-Net) to improve the network performance. In addition, we analyze the convergence of the Milstein scheme and theoretically guarantee the feasibility of MDSDE-Net. Experimental and theoretical results show that the MDSDE-Net outperforms existing models.
KW - Milstein-driven SDE-net
KW - Stochastic differential equation
KW - Uncertainty quantification
UR - http://www.scopus.com/inward/record.url?scp=85170651262&partnerID=8YFLogxK
U2 - 10.1016/j.patrec.2023.08.018
DO - 10.1016/j.patrec.2023.08.018
M3 - 文章
AN - SCOPUS:85170651262
SN - 0167-8655
VL - 174
SP - 71
EP - 77
JO - Pattern Recognition Letters
JF - Pattern Recognition Letters
ER -