Hypercomplex-valued neural networks, including quaternion-valued neural networks, can treat multi-dimensional data as a single entity. In this paper, we present the quaternion-valued recurrent projection neural networks (QRPNNs). Briefly, the QRPNNs are obtained by combining the non-local projection learning with the quaternion-valued recurrent correlation neural network (QRCNNs). We show that the QRPNNs overcome the cross-talk problem of the QRCNNs. Thus, they are appropriate to implement associative memories. Furthermore, computational experiments reveal that the QRPNNs exhibit greater storage capacity and noise tolerance than their corresponding QRCNNs.

超复杂值神经网络（包括四元数值神经网络）可以将多维数据视为单个实体。在本文中，我们提出了四元数值递归投影神经网络（QRPNNs）。简而言之，QRPNN是通过将非局部投影学习与四元数递归相关神经网络（QRCNN）相结合而获得的。我们表明，QRPNNs克服了QRCNNs的串扰问题。因此，它们适合于实现关联存储器。此外，计算实验表明，QRPNN比其相应的QRCNN具有更大的存储容量和噪声容忍度。