Baldi P F, Hornik K
Div. of Biol., California Inst. of Technol., Pasadena, CA.
IEEE Trans Neural Netw. 1995;6(4):837-58. doi: 10.1109/72.392248.
Networks of linear units are the simplest kind of networks, where the basic questions related to learning, generalization, and self-organization can sometimes be answered analytically. We survey most of the known results on linear networks, including: 1) backpropagation learning and the structure of the error function landscape, 2) the temporal evolution of generalization, and 3) unsupervised learning algorithms and their properties. The connections to classical statistical ideas, such as principal component analysis (PCA), are emphasized as well as several simple but challenging open questions. A few new results are also spread across the paper, including an analysis of the effect of noise on backpropagation networks and a unified view of all unsupervised algorithms.
线性单元网络是最简单的网络类型，在这类网络中，与学习、泛化和自组织相关的基本问题有时可以通过解析的方式得到解答。我们综述了线性网络的大部分已知成果，包括：1）反向传播学习和误差函数曲面的结构，2）泛化的时间演化，以及3）无监督学习算法及其性质。文中强调了与经典统计思想（如主成分分析（PCA））的联系，以及几个简单但具有挑战性的开放性问题。本文还散布了一些新成果，包括对噪声对反向传播网络影响的分析以及对所有无监督算法的统一观点。
