A structural optimization algorithm with stochastic forces and stresses

Abstract

We propose an algorithm for optimizations in which the gradients contain stochastic noise. This arises, for example, in structural optimizations when computations of forces and stresses rely on methods involving Monte Carlo sampling, such as quantum Monte Carlo or neural network states, or are performed on quantum devices which have intrinsic noise. Our proposed algorithm is based on the combination of two key ingredients: an update rule derived from the steepest descent method, and a staged scheduling of the targeted statistical error and step-size, with position averaging. We compare it with commonly applied algorithms, including some of the latest machine learning optimization methods, and show that the algorithm consistently performs efficiently and robustly under realistic conditions. Applying this algorithm, we achieve full-degree optimizations in solids using ab initio many-body computations, by auxiliary-field quantum Monte Carlo with planewaves and pseudopotentials. A new metastable structure in Si was discovered in a mixed geometry and lattice relaxing simulation. In addition to structural optimization in materials, our algorithm can potentially be useful in other problems in various fields where optimization with noisy gradients is needed.

arXiv comment18 pages, 14 figures

References in corpus (15)

• Escaping free-energy minima
• ADADELTA: An Adaptive Learning Rate Method
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Perspective on density functional theory
• Towards Quantum Chemistry on a Quantum Computer
• Quantum Monte Carlo method using phase-free random walks with Slater determinants
• Unbiasing Fermionic Quantum Monte Carlo with a Quantum Computer
• Zero-Variance Zero-Bias Principle for Observables in quantum Monte Carlo: Application to Forces
• Quantum Neural Network States: A Brief Review of Methods and Applications
• Phaseless auxiliary-field quantum Monte Carlo calculations with planewaves and pseudopotentials--applications to atoms and molecules
• Accurate, efficient and simple forces with Quantum Monte Carlo methods
• Coupled Electron-Ion Monte Carlo simulation of hydrogen molecular crystals
• High pressure hydrogen by machine learning and quantum Monte Carlo
• Optimized structure and vibrational properties by error affected potential energy surfaces
• Quantum Monte Carlo for minimum energy structures

Cited by in corpus (3)

• Computation of forces and stresses in solids: Towards accurate structural optimization with auxiliary-field quantum Monte Carlo
• Calculation of the energies of the multideterminant states of the nitrogen vacancy center in diamond with quantum Monte Carlo
• Deep learning quantum Monte Carlo for solids