In All Likelihood, Deep Belief Is Not Enough
In All Likelihood, Deep Belief Is Not Enough
复制标题
DOI：
10.5555/1953048.2078204
复制
发表时间：
2010-11
期刊：
影响因子：
--
通讯作者：
Lucas Theis;S. Gerwinn;Fabian H Sinz;M. Bethge
中科院分区：
文献类型：
--
作者：
Lucas Theis;S. Gerwinn;Fabian H Sinz;M. Bethge
自然图像的统计模型为机器学习和计算神经科学领域的研究人员提供了一个重要的工具。对统计模型的性能进行定量评价和比较的标准度量是似然。近年来越来越受欢迎并应用于各种复杂数据的一类统计模型是由深度信念网络形成的。然而，由于其可能性的计算难以处理的性质，这些模型的分析往往局限于基于样本的定性分析。在这种情况下，本文引入了一种计算易于处理且易于应用于实践的深度信念网络似然一致性估计器。利用该估计量，我们定量地研究了自然图像补丁的深度信念网络，并将其性能与其他自然图像补丁模型的性能进行了比较。我们发现，即使是非常简单的混合模型，深度信念网络在可能性方面也表现得更好。
Statistical models of natural images provide an important tool for researchers in the fields of machine learning and computational neuroscience. The canonical measure to quantitatively assess and compare the performance of statistical models is given by the likelihood. One class of statistical models which has recently gained increasing popularity and has been applied to a variety of complex data is formed by deep belief networks. Analyses of these models, however, have often been limited to qualitative analyses based on samples due to the computationally intractable nature of their likelihood. Motivated by these circumstances, the present article introduces a consistent estimator for the likelihood of deep belief networks which is computationally tractable and simple to apply in practice. Using this estimator, we quantitatively investigate a deep belief network for natural image patches and compare its performance to the performance of other models for natural image patches. We find that the deep belief network is outperformed with respect to the likelihood even by very simple mixture models.