The stabilizer formalism allows a subset of quantum circuits, the Clifford circuits made up of Hadamard, CNOT and phase gates, to be efficiently simulated on classical computers. However, universal computation is only possible by adding T gates to the set of simulated operations. The CAMPS (Clifford-augmented matrix product state) ansatz for quantum states joins the computational power of the stabilizer formalism and matrix product states, but in a naive implementation still struggles with simulating T gates due to an exponential increase in the reuqired bond dimension.
This project, a stepping stone to my Master's thesis, implements an analytical disentangling algorithm that keeps the MPS bond dimension to 1 for the first N simulated T gates, as described in the papers by Fux et al. (2025) and by Liu and Clark (2025).
G. E. Fux, B. Béri, R. Fazio, and E. Tirrito, “Disentangling Magic States with Classically Simulable Quantum Circuits,” Phys. Rev. Lett., vol. 135, no. 26, Dec. 2025, doi: 10.1103/ggp1-byj1.
Z. Liu and B. K. Clark, “Classical simulability of Clifford+T circuits with Clifford-augmented matrix product states,” Aug. 26, 2025, arXiv:2412.17209. doi: 10.48550/arXiv.2412.17209.