AI and the World's Hard Problems · Quantum physics & quantum computing · arXiv · 2026-10-23
Meta-learning of Gibbs states for many-body Hamiltonians with applications to Quantum Boltzmann Machines
The paper tackles preparing Gibbs (thermal) states of parametrized Hamiltonians on NISQ devices. The authors propose two meta-learning methods that train on a finite set of parameter values and then prepare states at unseen values within that range. The models are accurate on Ising and Heisenberg test cases up to eight qubits, and when applied to a Quantum Boltzmann Machine they cut runtime by 30 times compared with a VarQITE-based approach.