AUTO-ASSOCIATION BY MULTILAYER PERCEPTRONS AND SINGULAR VALUE DECOMPOSITION
AUTO-ASSOCIATION BY MULTILAYER PERCEPTRONS AND SINGULAR VALUE DECOMPOSITION
复制标题
DOI：
10.1007/bf00332918
复制
发表时间：
1988-01-01
影响因子：
1.9
通讯作者：
KAMP, Y
中科院分区：
文献类型：
--
作者：
BOURLARD, H;KAMP, Y
多层感知器在自动联想模式下工作时，有时被认为是信息处理应用中执行数据压缩或特征空间降维的有趣候选者。本论文表明，对于自关联，隐藏单元的非线性是无用的，并且最佳参数值可以直接由依赖于奇异值分解和低秩矩阵近似的纯线性技术导出，类似于著名的Karhunen-Loeve变换的精神。因此，这种方法似乎是一种有效的替代一般误差反向传播算法通常用于训练多层感知器。此外，它还对不同参数的作用给出了明确的解释。
The multilayer perceptron, when working in auto-association mode, is sometimes considered as an interesting candidate to perform data compression or dimensionality reduction of the feature space in information processing applications. The present paper shows that, for auto-association, the nonlinearities of the hidden units are useless and that the optimal parameter values can be derived directly by purely linear techniques relying on singular value decomposition and low rank matrix approximation, similar in spirit to the well-known Karhunen-Loeve transform. This approach appears thus as an efficient alternative to the general error back-propagation algorithm commonly used for training multilayer perceptrons. Moreover, it also gives a clear interpretation of the role of the different parameters.