Scalable Variational Ansatz for Quantum Many-Body Dynamics on Noisy Quantum Devices*

Quantum algorithms designed to simulate many-body quantum dynamics and their implementation on present-day noisy intermediate-scale quantum (NISQ) hardware should prioritize resource efficiency. Simulating time-dependent quantum Hamiltonians by explicit construction of the evolution operator requires a circuit depth scaling with the number of time steps, making it prohibitive for NISQ devices. Variational hybrid quantum-classical approaches, although approximate, make repeated use of short circuits and are thus more attractive for use in NISQ settings. However, finding a scalable variational ansatz capable of representing the dynamics in both adiabatic and non-adiabatic limits is a non-trivial problem. In our work, we use a variational time-evolution algorithm to simulate time-dependent spin models. Our ansatz scales polynomially with system size and provides qualitative agreement with exact numerics in adiabatic and non-adiabatic regimes with possible implementation in NISQ devices.