More Like this

1. Preparing quantum many-body scar states on quantum computers

   https://doi.org/10.22331/q-2023-11-07-1171  

   Gustafson, Erik J.; Li, Andy C.; Khan, Abid; Kim, Joonho; Kurkcuoglu, Doga Murat; Alam, M. Sohaib; Orth, Peter P.; Rahmani, Armin; Iadecola, Thomas (November 2023, Quantum)

   Quantum many-body scar states are highly excited eigenstates of many-body systems that exhibit atypical entanglement and correlation properties relative to typical eigenstates at the same energy density. Scar states also give rise to infinitely long-lived coherent dynamics when the system is prepared in a special initial state having finite overlap with them. Many models with exact scar states have been constructed, but the fate of scarred eigenstates and dynamics when these models are perturbed is difficult to study with classical computational techniques. In this work, we propose state preparation protocols that enable the use of quantum computers to study this question. We present protocols both for individual scar states in a particular model, as well as superpositions of them that give rise to coherent dynamics. For superpositions of scar states, we present both a system-size-linear depth unitary and a finite-depth nonunitary state preparation protocol, the latter of which uses measurement and postselection to reduce the circuit depth. For individual scarred eigenstates, we formulate an exact state preparation approach based on matrix product states that yields quasipolynomial-depth circuits, as well as a variational approach with a polynomial-depth ansatz circuit. We also provide proof of principle state-preparation demonstrations on superconducting quantum hardware. 

   more » « less
2. Scalable neural quantum states architecture for quantum chemistry

   https://doi.org/10.1088/2632-2153/acdb2f  

   Zhao, Tianchen; Stokes, James; Veerapaneni, Shravan (June 2023, Machine Learning: Science and Technology)

   Abstract Variational optimization of neural-network representations of quantum states has been successfully applied to solve interacting fermionic problems. Despite rapid developments, significant scalability challenges arise when considering molecules of large scale, which correspond to non-locally interacting quantum spin Hamiltonians consisting of sums of thousands or even millions of Pauli operators. In this work, we introduce scalable parallelization strategies to improve neural-network-based variational quantum Monte Carlo calculations forab-initioquantum chemistry applications. We establish GPU-supported local energy parallelism to compute the optimization objective for Hamiltonians of potentially complex molecules. Using autoregressive sampling techniques, we demonstrate systematic improvement in wall-clock timings required to achieve coupled cluster with up to double excitations baseline target energies. The performance is further enhanced by accommodating the structure of resultant spin Hamiltonians into the autoregressive sampling ordering. The algorithm achieves promising performance in comparison with the classical approximate methods and exhibits both running time and scalability advantages over existing neural-network based methods. 

   more » « less
3. Quasi-Monte Carlo Quasi-Newton in Variational Bayes

   Liu, Sifan; Owen, Art B. (January 2021, Journal of machine learning research)

   Many machine learning problems optimize an objective that must be measured with noise. The primary method is a first order stochastic gradient descent using one or more Monte Carlo (MC) samples at each step. There are settings where ill-conditioning makes second order methods such as limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) more effective. We study the use of randomized quasi-Monte Carlo (RQMC) sampling for such problems. When MC sampling has a root mean squared error (RMSE) of O(n−1/2) then RQMC has an RMSE of o(n−1/2) that can be close to O(n−3/2) in favorable settings. We prove that improved sampling accuracy translates directly to improved optimization. In our empirical investigations for variational Bayes, using RQMC with stochastic quasi-Newton method greatly speeds up the optimization, and sometimes finds a better parameter value than MC does. 

   more » « less
4. Estimating Bethe roots with VQE

   https://doi.org/10.1088/1751-8121/ad6db2  

   Raveh, David; Nepomechie, Rafael_I (August 2024, Journal of Physics A: Mathematical and Theoretical)

   Abstract Bethe equations, whose solutions determine exact eigenvalues and eigenstates of corresponding integrable Hamiltonians, are generally hard to solve. We implement a Variational Quantum Eigensolver approach to estimating Bethe roots of the spin-1/2 XXZ quantum spin chain, by using Bethe states as trial states, and treating Bethe roots as variational parameters. In numerical simulations of systems of size up to 6, we obtain estimates for Bethe roots corresponding to both ground states and excited states with up to 5 down-spins, for both the closed and open XXZ chains. This approach is not limited to real Bethe roots. 

   more » « less
5. Variational Marginal Particle Filters

   Lai, Jinlin; Domke, Justin; Sheldon, Daniel (January 2022, Proceedings of The International Conference on Artificial Intelligence and Statistics (AISTATS))

   Variational inference for state space models (SSMs) is known to be hard in general. Recent works focus on deriving variational objectives for SSMs from unbiased sequential Monte Carlo estimators. We reveal that the marginal particle filter is obtained from sequential Monte Carlo by applying Rao-Blackwellization operations, which sacrifices the trajectory information for reduced variance and differentiability. We propose the variational marginal particle filter (VMPF), which is a differentiable and reparameterizable variational filtering objective for SSMs based on an unbiased estimator. We find that VMPF with biased gradients gives tighter bounds than previous objectives, and the unbiased reparameterization gradients are sometimes beneficial. 

   more » « less