Towards Improved Quantum Machine Learning for Molecular Force Fields

Abstract

This study explores the use of equivariant quantum neural networks (QNN) for generating molecular force fields, focusing on the rMD17 dataset. We consider a QNN architecture based on previous research and point out shortcomings in the parametrization of the atomic environments. These shortcomings limit its expressivity as an interatomic potential and precludes transferability between molecules. We propose a revised QNN architecture that addresses these shortcomings. While both QNNs show promise in force prediction, with the revised architecture showing improved accuracy, they struggle with energy prediction. Further, both QNNs architectures fail to demonstrate a meaningful scaling law of decreasing errors with increasing training data. These findings highlight the challenges of scaling QNNs for complex molecular systems and emphasize the need for improved encoding strategies, regularization techniques, and hybrid quantum-classical approaches.

15 pages, 10 figures, Accepted in Physical Review A

References in corpus (37)

• Quantum algorithm for solving linear systems of equations
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• On representing chemical environments
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Quantum machine learning in feature Hilbert spaces
• Moment Tensor Potentials: a class of systematically improvable interatomic potentials
• Machine Learning of Accurate Energy-Conserving Molecular Force Fields
• Circuit-centric quantum classifiers
• The effect of data encoding on the expressive power of variational quantum machine learning models
• Challenges and Opportunities in Quantum Machine Learning
• On-the-fly machine learning force field generation: Application to melting points
• Phase transitions of hybrid perovskites simulated by machine-learning force fields trained on-the-fly with Bayesian inference
• Data re-uploading for a universal quantum classifier
• Quantum advantage in learning from experiments
• Polynomial-time quantum algorithm for the simulation of chemical dynamics
• An initialization strategy for addressing barren plateaus in parametrized quantum circuits
• Exploiting symmetry in variational quantum machine learning
• Optimizing many-body atomic descriptors for enhanced computational performance of machine learning based interatomic potentials
• Efficient nonparametric $n$-body force fields from machine learning
• Approximate amplitude encoding in shallow parameterized quantum circuits and its application to financial market indicator
• Calculating energy derivatives for quantum chemistry on a quantum computer
• Theory for Equivariant Quantum Neural Networks
• Theoretical Guarantees for Permutation-Equivariant Quantum Neural Networks
• Does provable absence of barren plateaus imply classical simulability?
• Understanding quantum machine learning also requires rethinking generalization
• Quantum Algorithm for Molecular Properties and Geometry Optimization
• Generalization despite overfitting in quantum machine learning models
• Efficient quantum computation of molecular forces and other energy gradients
• Provably Trainable Rotationally Equivariant Quantum Machine Learning
• Overfitting in quantum machine learning and entangling dropout
• Linear-depth quantum circuits for loading Fourier approximations of arbitrary functions
• Quantum Gaussian process model of potential energy surface for a polyatomic molecule
• Quantum Extreme Learning of molecular potential energy surfaces and force fields
• Symmetry-invariant quantum machine learning force fields
• Improved precision scaling for simulating coupled quantum-classical dynamics
• Encoded probabilistic imaginary-time evolution on a trapped-ion quantum computer for ground and excited states of spin qubits
• Benchmarking of quantum fidelity kernels for Gaussian process regression