Learning the Structure of Deep Sparse Graphical Models
Learning the Structure of Deep Sparse Graphical Models
复制标题
DOI：
--
复制
发表时间：
2009-12
期刊：
影响因子：
--
通讯作者：
Ryan P. Adams;Hanna M. Wallach;Zoubin Ghahramani
中科院分区：
文献类型：
--
作者：
Ryan P. Adams;Hanna M. Wallach;Zoubin Ghahramani
深度信念网络是模拟复杂概率分布的一种强大方法。然而，很难学习一个信念网络的结构，特别是一个隐藏的单位。印度自助过程已被用作一个单一的无限宽的隐藏层的有向信念网络的结构上的非参数贝叶斯先验。在这里，我们介绍了级联印度自助餐过程（CIBP），它提供了一个分层的，有向的信念网络的结构，是无限的深度和宽度，但允许听话的推理先验。我们使用CIBP先验与非线性高斯信念网络框架，允许每个单元在离散和连续表示之间改变其行为。我们在这个模型中使用马尔可夫链蒙特卡罗进行推理，并探索在图像数据上学习的结构。
Deep belief networks are a powerful way to model complex probability distributions. However, it is difficult to learn the structure of a belief network, particularly one with hidden units. The Indian buffet process has been used as a nonparametric Bayesian prior on the structure of a directed belief network with a single infinitely wide hidden layer. Here, we introduce the cascading Indian buffet process (CIBP), which provides a prior on the structure of a layered, directed belief network that is unbounded in both depth and width, yet allows tractable inference. We use the CIBP prior with the nonlinear Gaussian belief network framework to allow each unit to vary its behavior between discrete and continuous representations. We use Markov chain Monte Carlo for inference in this model and explore the structures learned on image data.