论文标题
用于密度估计的三角网络
A Triangular Network For Density Estimation
论文作者
论文摘要
我们报告了神经自回归流（NAF）的三角神经网络实施。与许多通用自回旋密度模型不同，我们的设计是高度模块化的，参数经济性的，计算上有效的，适用于高维数据的密度估计。它在通用密度估计器的类别中实现了MNIST和CIFAR-10（分别约为1.1和3.7）上的最新量化指数（分别约为1.1和3.7）。
We report a triangular neural network implementation of neural autoregressive flow (NAF). Unlike many universal autoregressive density models, our design is highly modular, parameter economy, computationally efficient, and applicable to density estimation of data with high dimensions. It achieves state-of-the-art bits-per-dimension indices on MNIST and CIFAR-10 (about 1.1 and 3.7, respectively) in the category of general-purpose density estimators.
