no code implementations • 1 Mar 2023 • Yani Feng, Kejun Tang, Xiaoliang Wan, Qifeng Liao
We present a dimension-reduced KRnet map approach (DR-KRnet) for high-dimensional Bayesian inverse problems, which is based on an explicit construction of a map that pushes forward the prior measure to the posterior measure in the latent space.
no code implementations • 23 Feb 2023 • Yunyu Huang, Yani Feng, Qifeng Liao
Drawing on the streaming variational Bayes method, we introduce the TT format into Bayesian tensor decomposition methods for streaming data, and formulate posteriors of TT cores.
no code implementations • 22 Feb 2023 • Yingzhi Xia, Qifeng Liao, Jinglai Li
To address these challenges, we propose a novel approximation method for estimating the high-dimensional posterior distribution.
no code implementations • 9 Jan 2023 • Zhihang Xu, Yingzhi Xia, Qifeng Liao
Bayesian inverse problems are often computationally challenging when the forward model is governed by complex partial differential equations (PDEs).
no code implementations • 11 Jul 2022 • Junjie He, Zhihang Xu, Qifeng Liao
Currently, deep neural network based methods are actively developed for learning governing equations in unknown dynamic systems, but their efficiency can degenerate for switching systems, where structural changes exist at discrete time instants.
no code implementations • 20 Mar 2021 • Kejun Tang, Xiaoliang Wan, Qifeng Liao
In this paper we present an adaptive deep density approximation strategy based on KRnet (ADDA-KR) for solving the steady-state Fokker-Planck (F-P) equations.
no code implementations • 21 Oct 2020 • Yani Feng, Kejun Tang, Lianxing He, Pingqiang Zhou, Qifeng Liao
This work proposes a novel tensor train random projection (TTRP) method for dimension reduction, where pairwise distances can be approximately preserved.
no code implementations • 24 Sep 2019 • Ke Li, Kejun Tang, Tianfan Wu, Qifeng Liao
A state-of-the-art deep domain decomposition method (D3M) based on the variational principle is proposed for partial differential equations (PDEs).