|
因式分解的3路受限Boltzmann机器在自然图像建模中的应用
Factored 3-way restricted Boltzmann machines for modeling natural images |
|
| |
| 课程网址: |
http://videolectures.net/aistats2010_ranzato_f3wr/
|
| 主讲教师: |
Marc’Aurelio Ranzato
|
| 开课单位: |
多伦多大学
|
| 开课时间: |
2010-06-03
|
| 课程语种: |
英语
|
| 中文简介: |
深信网在手写体汉字建模方面取得了成功，但在实际图像中应用难度较大。问题出在受限玻尔兹曼机(restricted Boltzmann machine, RBM)上，RBM是用来学习深度信念网的模块，每次只学习一层。利用高斯-二值RBMs对实值数据进行建模并不能很好地模拟自然图像的协方差结构。我们提出了一种因式3路RBM，它使用其隐藏单元的状态来表示图像局部协方差结构中的异常。这为广泛使用的简单/复杂单元架构提供了一个概率框架。我们的模型学习了二进制特征，这些特征对于“微小图像”上的目标识别非常有效。通过使用标准的二进制RBM来学习更深层次的模型，可以得到更好的特征。
|
| 课程简介: |
Deep belief nets have been successful in modeling handwritten characters, but it has proved more difficult to apply them to real images. The problem lies in the restricted Boltzmann machine (RBM) which is used as a module for learning deep belief nets one layer at a time. The Gaussian-Binary RBMs that have been used to model real-valued data are not a good way to model the covariance structure of natural images. We propose a factored 3-way RBM that uses the states of its hidden units to represent abnormalities in the local covariance structure of an image. This provides a probabilistic framework for the widely used simple/complex cell architecture. Our model learns binary features that work very well for object recognition on the "tiny images" data set. Even better features are obtained by then using standard binary RBM's to learn a deeper model.
|
| 关 键 词: |
Boltzmann机器; 自然图像建模
|
| 课程来源: |
视频讲座网
|
| 最后编审: |
2021-01-30：nkq
|
| 阅读次数: |
31
|
|
图 书 馆