Crystal structure prediction using graph neural combinatorial optimization

• URL: http://arxiv.org/abs/2604.23921v1
• Date: Mon, 27 Apr 2026 00:34:03 GMT
• Title: Crystal structure prediction using graph neural combinatorial optimization
• Abstract summary: Crystalline materials are widely used in technological applications, yet their discovery remains a significant challenge.<n>As their properties are driven by structure, crystal structure prediction (CSP) methods play a central role in computational approaches aiming to accelerate this process.<n>We introduce a neural optimization approach to the atom allocation challenge and, subsequently, CSP, based on graph neural networks (GNNs)<n>We demonstrate that our method outperforms classical approaches and is competitive with a commercial optimization solver across a range of chemical compositions.
• Score: 2.656984940355681
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Crystalline materials are widely used in technological applications, yet their discovery remains a significant challenge. As their properties are driven by structure, crystal structure prediction (CSP) methods play a central role in computational approaches aiming to accelerate this process. Previously, CSP has been approached from a combinatorial optimization perspective, with the core challenge of allocating atoms on a fine grid of predefined discrete positions within a unit cell while minimizing their interaction energy. Exact mathematical optimization methods provide guaranteed solutions, but they become computationally expensive for large-scale instances, where the atomic configuration space grows rapidly, particularly in the absence of additional symmetry constraints. In this work, we introduce a neural combinatorial optimization approach to the atom allocation challenge and, subsequently, CSP, based on graph neural networks (GNNs), which can effectively sample from the distribution of feasible structures in an unsupervised manner. We leverage expander graphs to construct computational graphs over discrete positions that capture both short- and long-range interactions between atoms, and employ the Gumbel-Sinkhorn approach to enforce the desired stoichiometry of the generated structures. We demonstrate that our method outperforms classical heuristic approaches and is competitive with a commercial optimization solver across a range of chemical compositions. This enables the use of ever-expanding GPU infrastructure to tackle the inherent combinatorial challenges of CSP, paving the way for scaling beyond current capabilities.

Related papers

• An Iterative Dual-Channel Neural Quantum State Algorithm for Selected Configuration Interaction [3.013377814843082]
  We introduce the Handover Iterative Neural Quantum State (HI-NQS) algorithm, which embeds a classically trained autoregressive Transformer neural quantum state within the iterative sample--diagonalize--update framework of Sample-Based Quantum Diagonalization.<n>HI-NQS achieves chemical accuracy on all systems tested, with determinant-count scaling substantially more favorable than conventional CIPSI-based SCI for all but the smallest active spaces.
  arXiv  Detail & Related papers  (2026-06-25T08:47:04Z)
• Auto-regressive Neural Quantum State Sampling for Selected Configuration Interaction [0.0]
  We introduce an algorithm that uses auto-regressive neural networks (ARNNs) to guide subspace expansion for ground-state search.<n>Our algorithm efficiently constructs compact variational subspaces from learned ground-state statistics.<n> Benchmarks on molecular systems demonstrate that ARNN-guided subspace expansion combines the strengths of neural-network representations and classical subspace methods.
  arXiv  Detail & Related papers  (2026-03-25T18:53:00Z)
• Global Optimization of Atomic Clusters via Physically-Constrained Tensor Train Decomposition [37.69102815774395]
  We introduce a novel framework that overcomes the limitation by exploiting the low-rank structure of potential energy surfaces.<n>We demonstrate the efficacy of our method by identifying global minima of Lennard-Jones clusters containing up to 45 atoms.<n>We establish its practical applicability to real-world systems by optimizing 20-atom carbon clusters using a machine-learned Moment Potential.
  arXiv  Detail & Related papers  (2026-01-26T15:38:20Z)
• Leveraging Discrete Function Decomposability for Scientific Design [48.365465744654365]
  In the era of AI-driven science and engineering, we often want to design objects in silico according to user-specified properties.<n>For example, we may wish to design a protein to bind its target, arrange components within a circuit to minimize latency, or find materials with certain properties.<n>We propose and demonstrate use of a new distributional optimization algorithm, De-Aware Distributional Optimization (DADO), that can leverage any decomposability defined by a junction tree on the design variables.
  arXiv  Detail & Related papers  (2025-11-04T21:57:51Z)
• Quantum-Classical Hybrid Quantized Neural Network [8.382617481718643]
  We present a novel Quadratic Binary Optimization (QBO) model for quantized neural network training, enabling the use of arbitrary activation and loss functions.<n>We employ the Quantum Gradient Conditional Descent (QCGD) algorithm, which leverages quantum computing to directly solve the QCBO problem.
  arXiv  Detail & Related papers  (2025-06-23T02:12:36Z)
• A quantum wire approach to weighted combinatorial graph optimisation problems [0.0]
  We present and experimentally demonstrate an efficient encoding scheme based on chains of Rydberg-blockaded atoms.<n>We embed maximum weighted independent set (MWIS) and quadratic unconstrained binary optimization (QUBO) problems on a neutral atom architecture.
  arXiv  Detail & Related papers  (2025-03-21T13:00:51Z)
• Neutron-nucleus dynamics simulations for quantum computers [49.369935809497214]
  We develop a novel quantum algorithm for neutron-nucleus simulations with general potentials. It provides acceptable bound-state energies even in the presence of noise, through the noise-resilient training method. We introduce a new commutativity scheme called distance-grouped commutativity (DGC) and compare its performance with the well-known qubit-commutativity scheme.
  arXiv  Detail & Related papers  (2024-02-22T16:33:48Z)
• Latent Conservative Objective Models for Data-Driven Crystal Structure Prediction [62.36797874900395]
  In computational chemistry, crystal structure prediction is an optimization problem. One approach to tackle this problem involves building simulators based on density functional theory (DFT) followed by running search in simulation. We show that our approach, dubbed LCOMs (latent conservative objective models), performs comparably to the best current approaches in terms of success rate of structure prediction.
  arXiv  Detail & Related papers  (2023-10-16T04:35:44Z)
• Automatic and effective discovery of quantum kernels [41.61572387137452]
  Quantum computing can empower machine learning models by enabling kernel machines to leverage quantum kernels for representing similarity measures between data.<n>We present an approach to this problem, which employs optimization techniques, similar to those used in neural architecture search and AutoML.<n>The results obtained by testing our approach on a high-energy physics problem demonstrate that, in the best-case scenario, we can either match or improve testing accuracy with respect to the manual design approach.
  arXiv  Detail & Related papers  (2022-09-22T16:42:14Z)
• Scalable neural quantum states architecture for quantum chemistry [5.603379389073144]
  Variational optimization of neural-network representations of quantum states has been successfully applied to solve interacting fermionic problems. We introduce scalable parallelization strategies to improve neural-network-based variational quantum Monte Carlo calculations for ab-initio quantum chemistry applications.
  arXiv  Detail & Related papers  (2022-08-11T04:40:02Z)
• Tree ensemble kernels for Bayesian optimization with known constraints over mixed-feature spaces [54.58348769621782]
  Tree ensembles can be well-suited for black-box optimization tasks such as algorithm tuning and neural architecture search. Two well-known challenges in using tree ensembles for black-box optimization are (i) effectively quantifying model uncertainty for exploration and (ii) optimizing over the piece-wise constant acquisition function. Our framework performs as well as state-of-the-art methods for unconstrained black-box optimization over continuous/discrete features and outperforms competing methods for problems combining mixed-variable feature spaces and known input constraints.
  arXiv  Detail & Related papers  (2022-07-02T16:59:37Z)
• Efficient Micro-Structured Weight Unification and Pruning for Neural Network Compression [56.83861738731913]
  Deep Neural Network (DNN) models are essential for practical applications, especially for resource limited devices. Previous unstructured or structured weight pruning methods can hardly truly accelerate inference. We propose a generalized weight unification framework at a hardware compatible micro-structured level to achieve high amount of compression and acceleration.
  arXiv  Detail & Related papers  (2021-06-15T17:22:59Z)
• Adaptive pruning-based optimization of parameterized quantum circuits [62.997667081978825]
  Variisy hybrid quantum-classical algorithms are powerful tools to maximize the use of Noisy Intermediate Scale Quantum devices. We propose a strategy for such ansatze used in variational quantum algorithms, which we call "Efficient Circuit Training" (PECT) Instead of optimizing all of the ansatz parameters at once, PECT launches a sequence of variational algorithms.
  arXiv  Detail & Related papers  (2020-10-01T18:14:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.