Distributed-Memory Tensor Network Algorithms for Simulating Open Quantum Systems

Open quantum systems, typically governed by the Linblad quantum master equation (LQME), are a useful model for simulating a real quantum computer with noise. We investigate the non-equilibrium dynamics of open quantum systems under the LQME using tensor network states. Using the Cyclops Tensor Framework (CTF), a distributed memory tensor contraction library, we simulate the evolution of a quantum system by representing the density matrix as a matrix product state (MPS) under a Choi isomorphism. The massively parallel nature of our algorithm will let us perform tensor operations at a faster rate than traditional libraries. This will allow us to use larger tensors, thus obtaining a more accurate ansatz. We will use this simulation to gauge the fidelity of noisy quantum computers more efficiently.