A Quantum-Inspired Machine Learning Based on the Heisenberg Uncertainty

A quantum-inspired machine learning framework based on the Heisenberg uncertainty principle is presented in this article. First, we will give the geometric interpretation of the Heisenberg uncertainty principle. Second, the derived uncertainty relation is used to construct a machine learning model, named Heisenberg Bases. The computational experiments demonstrate that the proposed model's performance is comparable to that of the conventional FNN, GRN, RNN, and Transformer, as shown in three case studies. Moreover, a neural-based cell automata created from the Heisenberg bases is introduced and demonstrated in two use cases.