1: \begin{abstract}

2: 	 The multiple-valued quantum logic is formulated in a systematic way using the Bargmann representation of quantum basis states. In this approach, the truth values are represented naturally as  unique roots of unity placed on the unit circle.  Consequently, multi-valued quantum neurons are based on the principles of multiple-valued threshold logic over the field of complex numbers.  The training of MVQN is reduced to the movement along the unit circle. A quantum neural networks (QNNs)  based on multi-valued quantum neurons can be constructed with complex weights, inputs, and  outputs encoded by roots of unity and an activation function that  maps the complex plane into the unit circle. Such neural networks enjoy fast convergence and higher functionalities compared with quantum neural networks based on binary input with the same number of neurons and layers. Possible practical application can be found using the orbital angular momentum (OAM) of light based QNNs. 

3: 	\end{abstract}

4: