Generation and Reconstruction of large-amplitude Cat States Based on Squeezed Magnon-Qubit hybrid systems

As a bridge connecting different quantum systems, magnons have demonstrated exceptional capabilities in coherent coupling with various quantum systems such as photons, phonons, and qubits. As a result, hybrid systems composed of magnons have made significant progress in quantum technologies, quantum networks, and the construction of resource quantum states. Cat states, especially large-amplitude cat states, as a kind of important resource states, are not only important tools for testing quantum theory but also a cornerstone in the construction of logical qubits. However, large cat states in magnon systems are still challenging to prepare. A possible solution can be quantum state reconstruction using neural networks, which can partially transform the challenges in manipulating quantum systems into routine tasks in machine learning. We investigate a hybrid system consisting of squeezed magnons and superconducting flux qubits. This system enables enhanced generation of large Schrodinger cat states. Furthermore, we propose a method using neural ff network-based quantum state reconstruction to address two adverse factors: dissipation in magnon cat states and noise generated during measurements. Our research provides a feasible scheme for the preparation of large-amplitude cat states and offers viable approaches for the acquisition and error correction of quantum information in magnon cat states.