Journal of Modern Physics

Machine Learning for Quantum Rotor Dynamics: A Surrogate Model Approach to Rotational Excitation in External Electric Fields

Vinod Prasad, Ashish Tyagi, Urvashi Arya, Brijender Dahiya

Published Jan 1, 2026• 0 citations

Abstract

The quantum rotor in external fields such as static electric, magnetic, and time-dependent fields is one of the most studied research problems due to its applicability in chemical physics, molecular dynamics, and many other related areas. The problem finds applications in quantum chaos and nonlinear dynamics as well. In this work, we present a Machine learning (ML) approach to model a one-dimensional quantum rotor in an external electric field. Our approach is to develop a surrogate model, keeping in mind the inherent smoothness of the eigenenergies of the dressed rotor and the directional parameters such as orientation, i.e., expectation value of cosϕ , and the alignment parameter ( 〈 cos 2 ϕ 〉 ) with the applied static field. The model we developed is very efficient, achieving microsecond-scale predictions with high accuracy. The model we developed is as follows: first, we train it on the basis of data generated from numerical calculations taking rotor-static field interaction into account and expanding the wavefunction as a finite basis. The model gives an excellent performance with R2 values exceeding 0.9999 for all predicted quantities. We further explore the physics of the system and the data generation, the details of the ML model, and its uses in designing an inverse model. The work shows that ML can be used to speed up computational time in related areas such as molecular physics and may be used to optimize the experimental parameters. This approach is directly extensible to three-dimensional rotors, time-dependent fields, and multi-parameter optimization, with potential applications in quantum control, molecular alignment, and spectroscopic fitting.

DOI: 10.4236/jmp.2026.177037

Related papers

Advancements in predictive modeling of nuclear magnetic resonance parameters: integrating quantum mechanics, machine learning, and quantum computing

Jonhariono Sihotang, Patrisius Michaud Felix Marsoit

This research explores the integration of quantum mechanics, machine learning, and quantum computing to advance predictive modeling of nuclear magnetic resonance (NMR) parameters. The aim is to develop a hybrid quantum-enhanced machine learning model that combines the accuracy of quantum calculations with the efficiency of machine learning techniques for predicting NMR chemical shifts. The conceptual framework involves quantum mechanical calculations for accurate reference NMR parameters, supervised machine learning models trained on diverse molecular datasets, and hybrid quantum-classical algorithms to leverage quantum computing resources. A simplified numerical example demonstrates the potential of the proposed model for predicting NMR chemical shifts for small molecular systems. The results showcase the model's ability to capture underlying relationships between molecular features and NMR observables, indicating promise for larger and more complex systems. This interdisciplinary approach opens new avenues for advancing NMR spectroscopy and understanding molecular structures, dynamics, and interactions in various scientific domains. The research also discusses challenges and opportunities in integrating quantum mechanics, machine learning, and quantum computing, emphasizing the importance of diverse datasets and quantum algorithm selection. The proposed model holds significant implications for transforming NMR parameter predictions and contributing to chemistry, biochemistry, and materials science research

VertexDec 30, 20220 citations

Title: Non-Linear Quantum Field Theory with Adaptive Parameterization

Jincheng Zhang

This paper introduces a novel non-linear quantum field theory framework designed to enhance model accuracy and computational efficiency. Traditional field theories frequently rely on static, predefined parameters, limiting their applicability to specific physical scenarios. We propose a dynamic parameterization system that iteratively adjusts these parameters based on model performance, creating a feedback loop that continuously refines the model's fit. The core mechanism centers around a reinforcement learning-inspired feedback loop, evaluating the model's output against new data and optimizing parameter values to maximize predictive power. This approach overcomes the limitations of fixed parameter models, offering a more flexible and efficient framework for simulating complex quantum phenomena. The resulting model demonstrates improved performance across a range of benchmark problems, highlighting the potential for this adaptive approach in diverse physical contexts.

Sep 1, 20260 citations

SYNERGY BETWEEN QUANTUM COMPUTING AND MACHINE LEARNING IN QUANTUM NEURAL NETWORK

Sadia Rafi, Rana Muhammad Ali Maisam, Anum Zahid +3 more

Machine learning has made significant contributions to the fields of chemistry and materials science, enabling the exploration of vast chemical space through large-scale quantum chemical calculations. These models provide fast and accurate predictions of atomistic chemical properties, but they have limitations when it comes to capturing the electronic degrees of freedom of a molecule. This restricts their application in reactive chemistry and chemical analysis. To address this limitation, we introduce a deep learning framework that predicts the quantum mechanical wavefunction of a molecule in a local basis of atomic orbitals. The wavefunction serves as a foundational representation from which all other ground-state properties can be derived. Our approach maintains complete access to the electronic structure through the wavefunction, while achieving computational efficiency comparable to force-field methods. Moreover, the framework captures quantum mechanics in a form that can be analytically differentiated, allowing for efficient optimization and exploration of chemical systems. We demonstrate the potential of our approach through several examples. By leveraging the predicted wavefunction, we showcase the ability to perform inverse design of molecular structures to target specific electronic property optimizations. This opens exciting avenues for tailoring molecular structures to achieve desired electronic characteristics. Additionally, our framework paves the way for enhanced synergy between machine learning and quantum chemistry, enabling more comprehensive investigations into complex chemical systems.

CENTRAL ASIAN JOURNAL OF MATHEMATICAL THEORY AND COMPUTER SCIENCESAug 28, 20230 citations

Machine Learning Many-Body Green’s Functions for Molecular Excitation Spectra

Christian Venturella, Christopher Hillenbrand, Jiachen Li +1 more

We present a machine learning (ML) framework for predicting Green’s functions of molecular systems, from which photoemission spectra and quasiparticle energies at quantum many-body level can be obtained. Kernel ridge regression is adopted to predict self-energy matrix elements on compact imaginary frequency grids from static and dynamical mean-field electronic features, which gives direct access to real-frequency many-body Green’s functions through analytic continuation and Dyson’s equation. Feature and self-energy matrices are represented in a symmetry-adapted intrinsic atomic orbital plus projected atomic orbital basis to enforce rotational invariance. We demonstrate good transferability and high data efficiency of the proposed ML method across molecular sizes and chemical species by showing accurate predictions of density of states (DOS) and quasiparticle energies at the level of many-body perturbation theory ( GW ) or full configuration interaction. For the ML model trained on 48 out of 1995 molecules randomly sampled from the QM7 and QM9 data sets, we report the mean absolute errors of ML-predicted highest occupied and lowest unoccupied molecular orbital energies to be 0.13 and 0.10 eV, respectively, compared to GW @PBE0. We further showcase the capability of this method by applying the same ML model to predict DOS for significantly larger organic molecules with up to 44 heavy atoms.

arXiv:2310.09911Journal of Chemical Theory and ComputationDec 27, 202317 citations

Dimension Expansion Quantum Computing: Adaptive Learning Algorithm

Jincheng Zhang

Quantum computing offers unprecedented computational power, but scaling this power requires significant computational resources and careful parameter tuning. Traditional optimization techniques often rely on fixed parameters, limiting the potential for efficient scaling. This paper proposes a novel self-adaptive learning algorithm for quantum system parameter optimization, leveraging machine learning to dynamically adjust parameters based on the quantum system's inherent properties – specifically, quantum spin and quantum entanglement. The algorithm's core mechanism involves training a model to predict optimal parameter values, which are then used to iteratively refine the system's state, ultimately enhancing quantum computation's scalability and performance. This work addresses the limitations of static parameter optimization by introducing a dynamic and adaptive approach that exploits the unique characteristics of quantum systems. The research explores the potential of this algorithm to significantly improve the efficiency and expandability of quantum computations.

Aug 31, 20260 citations

Quantum-centric machine learning for molecular dynamics

Yanxian Tao, Lingyun Wan, Xiongzhi Zeng +4 more

Accurate and efficient prediction of electronic wavefunctions is central to ab initio molecular dynamics (AIMD) and electronic structure theory. However, conventional ab initio methods require self-consistent optimization of electronic states at every nuclear configuration, leading to prohibitive computational costs, especially for large or strongly correlated systems. Here, we introduce a quantum-centric machine learning (QCML) model-a hybrid quantum-classical framework that integrates parameterized quantum circuits (PQCs) with Transformer-based machine learning to directly predict molecular wavefunctions and quantum observables. By pretraining the Transformer on a diverse dataset of molecules and ansatz types and subsequently fine-tuning it for specific systems, QCML learns transferable mappings between molecular descriptors and PQC parameters, eliminating the need for iterative variational optimization. The pretrained model achieves chemical accuracy in potential energy surfaces, atomic forces, and dipole moments across multiple molecules and ansatzes, and enables efficient AIMD simulations with infrared spectra prediction. This work establishes a scalable and transferable quantum-centric machine learning paradigm, bridging variational quantum algorithms and modern deep learning for next-generation molecular simulation and quantum chemistry applications.

physics.chem-phCross-listedarXiv:2511.07771Nov 12, 20251 citation