Deep density estimation via invertible block-triangular mapping
Deep density estimation via invertible block-triangular mapping
复制标题
DOI：
10.1016/j.taml.2020.01.023
复制
发表时间：
2020-03
影响因子：
3.4
通讯作者：
Keju Tang;X. Wan;Qifeng Liao
中科院分区：
文献类型：
--
作者：
Keju Tang;X. Wan;Qifeng Liao
在这项工作中，我们通过耦合Knothe-Rosenblatt （KR）重排和基于流的生成模型，开发了一个称为KRnet的可逆传输映射，用于密度估计，该模型推广了实值非体积保持（real NVP）模型（arX-iv:1605.08803v3）。KR重排的三角形结构打破了真实NVP在维度间信息交换方面的对称性，不仅加快了训练过程，而且显著提高了训练精度。我们还在生成模型中引入了几个新层，以提高鲁棒性和有效性，包括重新制定的仿射耦合层，旋转层和组件非线性可逆层。KRnet可以用于密度估计和样本生成，特别是在维数相对较高的情况下。通过数值实验验证了KRnet的性能。
In this work, we develop an invertible transport map, called KRnet, for density estimation by coupling the Knothe–Rosenblatt (KR) rearrangement and the flow-based generative model, which generalizes the real-valued non-volume preserving (real NVP) model (arX-iv:1605.08803v3). The triangular structure of the KR rearrangement breaks the symmetry of the real NVP in terms of the exchange of information between dimensions, which not only accelerates the training process but also improves the accuracy significantly. We have also introduced several new layers into the generative model to improve both robustness and effectiveness, including a reformulated affine coupling layer, a rotation layer and a component-wise nonlinear invertible layer. The KRnet can be used for both density estimation and sample generation especially when the dimensionality is relatively high. Numerical experiments have been presented to demonstrate the performance of KRnet.