Choi J Y, Choi C H
Dept. of Control and Instrum. Eng., Seoul Nat. Univ.
IEEE Trans Neural Netw. 1992;3(1):101-7. doi: 10.1109/72.105422.
In a neural network, many different sets of connection weights can approximately realize an input-output mapping. The sensitivity of the neural network varies depending on the set of weights. For the selection of weights with lower sensitivity or for estimating output perturbations in the implementation, it is important to measure the sensitivity for the weights. A sensitivity depending on the weight set in a single-output multilayer perceptron (MLP) with differentiable activation functions is proposed. Formulas are derived to compute the sensitivity arising from additive/multiplicative weight perturbations or input perturbations for a specific input pattern. The concept of sensitivity is extended so that it can be applied to any input patterns. A few sensitivity measures for the multiple output MLP are suggested. For the verification of the validity of the proposed sensitivities, computer simulations have been performed, resulting in good agreement between theoretical and simulation outcomes for small weight perturbations.
在神经网络中，许多不同的连接权重集都可以近似实现输入-输出映射。神经网络的灵敏度会因权重集的不同而变化。为了选择灵敏度较低的权重或在实现过程中估计输出扰动，测量权重的灵敏度非常重要。本文提出了一种依赖于具有可微激活函数的单输出多层感知器（MLP）中权重集的灵敏度。推导了用于计算特定输入模式下由加性/乘性权重扰动或输入扰动引起的灵敏度的公式。灵敏度的概念得到了扩展，以便可以应用于任何输入模式。还提出了一些针对多输出MLP的灵敏度度量。为了验证所提出灵敏度的有效性，进行了计算机模拟，结果表明对于小权重扰动，理论结果与模拟结果吻合良好。
