Reed R, Marks R J, Oh S
Dept. of Electr. Eng., Washington Univ., Seattle, WA.
IEEE Trans Neural Netw. 1995;6(3):529-38. doi: 10.1109/72.377960.
The generalization performance of feedforward layered perceptrons can, in many cases, be improved either by smoothing the target via convolution, regularizing the training error with a smoothing constraint, decreasing the gain (i.e., slope) of the sigmoid nonlinearities, or adding noise (i.e., jitter) to the input training data, In certain important cases, the results of these procedures yield highly similar results although at different costs. Training with jitter, for example, requires significantly more computation than sigmoid scaling.
在许多情况下，前馈分层感知器的泛化性能可以通过以下方式得到提升：通过卷积平滑目标、用平滑约束对训练误差进行正则化、降低Sigmoid非线性函数的增益（即斜率），或者向输入训练数据添加噪声（即抖动）。在某些重要情况下，这些方法的结果虽然成本不同，但却非常相似。例如，使用抖动进行训练比Sigmoid缩放需要更多的计算。
