Ridge functions are used to describe and study the lower bound of the approximation done by the neural networks which can be written as a linear combination of activation functions. If the activation functions are also ridge functions, these networks are called explainable neural networks. In this paper, we first show that quantum neural networks which are based on variational quantum circuits can be written as a linear combination of ridge functions. Consequently, we show that the interpretability and explainability of such quantum neural networks can be directly considered and studied as an approximation with the linear combination of ridge functions.
翻译：脊柱功能用于描述和研究神经网络近似值的下限,该近似值可以写成激活功能的线性组合。如果激活功能也是脊柱功能,则这些网络被称为可解释的神经网络。在本文中,我们首先表明,基于可变量子电路的量子神经网络可以写成脊柱功能的线性组合。因此,我们表明,可直接考虑并研究此类量子神经网络的可解释性和可解释性,将其作为与脊柱功能线性组合的近似值。
相关内容
微信扫码咨询专知VIP会员