arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1602.03264stat.MLcs.LG
收藏

生成ConvNet理论

A Theory of Generative ConvNet

University of California, Los Angeles(加州大学洛杉矶分校)

Jianwen Xie, Yang Lu, Song-Chun Zhu, Ying Nian Wu

摘要:

我们证明了一个生成的随机场模型,我们称之为生成的ConvNet,可以从常用的判别ConvNet中导出,通过假设多类别分类的ConvNet,并假设其中一个类别是由参考分布生成的基本类别。如果我们进一步假设ConvNet中的非线性是整流线性单元(ReLU),参考分布是高斯白噪声,那么我们得到一个在基于能量的模型中唯一的生成ConvNet模型:该模型是分段高斯的,并且高斯片段的均值由自动编码器定义,其中自下而上编码中的滤波器成为自上而下解码中的基函数,并且在自底向上卷积过程中由滤波器检测的二进制激活变量变成自顶向下去卷积过程中的基函数的系数。用于对生成ConvNet进行采样的Langevin动态由该自动编码器的重构误差驱动。生成式ConvNet的对比发散学习通过自动编码器重建训练图像。极大似然学习算法可以合成逼真的自然图像模式。

英文摘要:

We show that a generative random field model, which we call generative ConvNet, can be derived from the commonly used discriminative ConvNet, by assuming a ConvNet for multi-category classification and assuming one of the categories is a base category generated by a reference distribution. If we further assume that the non-linearity in the ConvNet is Rectified Linear Unit (ReLU) and the reference distribution is Gaussian white noise, then we obtain a generative ConvNet model that is unique among energy-based models: The model is piecewise Gaussian, and the means of the Gaussian pieces are defined by an auto-encoder, where the filters in the bottom-up encoding become the basis functions in the top-down decoding, and the binary activation variables detected by the filters in the bottom-up convolution process become the coefficients of the basis functions in the top-down deconvolution process. The Langevin dynamics for sampling the generative ConvNet is driven by the reconstruction error of this auto-encoder. The contrastive divergence learning of the generative ConvNet reconstructs the training images by the auto-encoder. The maximum likelihood learning algorithm can synthesize realistic natural image patterns.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑