Grand canonical generative diffusion model for crystalline phases and grain boundaries

• URL: http://arxiv.org/abs/2408.15601v1
• Date: Wed, 28 Aug 2024 07:49:29 GMT
• Title: Grand canonical generative diffusion model for crystalline phases and grain boundaries
• Authors: Bo Lei, Enze Chen, Hyuna Kwon, Tim Hsu, Babak Sadigh, Vincenzo Lordi, Timofey Frolov, Fei Zhou,
• Abstract summary: The diffusion model has emerged as a powerful tool for generating atomic structures for materials science. This work calls attention to the deficiency of current particle-based diffusion models, which represent atoms as a point cloud. We develop a solution, the grand canonical diffusion model, which adopts an alternative voxel-based representation with continuous rather than fixed number of particles.
• Score: 8.159060728081203
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The diffusion model has emerged as a powerful tool for generating atomic structures for materials science. This work calls attention to the deficiency of current particle-based diffusion models, which represent atoms as a point cloud, in generating even the simplest ordered crystalline structures. The problem is attributed to particles being trapped in local minima during the score-driven simulated annealing of the diffusion process, similar to the physical process of force-driven simulated annealing. We develop a solution, the grand canonical diffusion model, which adopts an alternative voxel-based representation with continuous rather than fixed number of particles. The method is applied towards generation of several common crystalline phases as well as the technologically important and challenging problem of grain boundary structures.

Related papers

• Fourier Transformers for Latent Crystallographic Diffusion and Generative Modeling [15.281839359113446]
  New crystalline materials call for generative models that handle periodic boundary conditions, crystallographic symmetries, and physical constraints.<n>We propose a reciprocal-space generative pipeline that represents crystals through a truncated Fourier transform of the species-resolved unit-cell density.
  arXiv  Detail & Related papers  (2026-02-12T15:11:12Z)
• C2NP: A Benchmark for Learning Scale-Dependent Geometric Invariances in 3D Materials Generation [2.3982445219832678]
  We introduce Crystal-to-Nanoparticle (C2NP), a benchmark for evaluating generative models when moving between infinite crystalline unit cells and finite nanoparticles.<n>C2NP defines two complementary tasks: (i) generating nanoparticles of specified radii from periodic unit cells, testing whether models capture surface truncation and geometric constraints; and (ii) recovering bulk lattice parameters and space-group symmetry from finite particle configurations.
  arXiv  Detail & Related papers  (2026-01-27T01:25:50Z)
• Kinetic Langevin Diffusion for Crystalline Materials Generation [8.275553926021168]
  We introduce Kinetic Langevin Diffusion for Materials (KLDM), a novel diffusion model for crystalline materials generation.<n>By coupling coordinates with auxiliary Euclidean variables representing velocities, the diffusion process is now offset to a flat space.<n>We evaluate KLDM on both Crystal Structure Prediction (CSP) and De-novo Generation (DNG) tasks.
  arXiv  Detail & Related papers  (2025-07-04T14:18:26Z)
• Critical Iterative Denoising: A Discrete Generative Model Applied to Graphs [52.50288418639075]
  We propose a novel framework called Iterative Denoising, which simplifies discrete diffusion and circumvents the issue by assuming conditional independence across time. Our empirical evaluations demonstrate that the proposed method significantly outperforms existing discrete diffusion baselines in graph generation tasks.
  arXiv  Detail & Related papers  (2025-03-27T15:08:58Z)
• Exploring the flavor structure of leptons via diffusion models [0.0]
  We consider a simple extension of the Standard Model with the type I seesaw mechanism and train a neural network to generate the neutrino mass matrix. By utilizing transfer learning, the diffusion model generates 104 solutions consistent with the neutrino mass squared differences and the leptonic mixing angles.
  arXiv  Detail & Related papers  (2025-03-27T12:17:00Z)
• A Periodic Bayesian Flow for Material Generation [20.62877413439857]
  We introduce CrysBFN, a novel crystal generation method by proposing a periodic Bayesian flow. To successfully realize the concept of periodic Bayesian flow, CrysBFN integrates a new entropy conditioning mechanism. Experiments over both crystal ab initio generation and crystal structure prediction tasks demonstrate the superiority of CrysBFN.
  arXiv  Detail & Related papers  (2025-02-04T05:07:13Z)
• Conditional Synthesis of 3D Molecules with Time Correction Sampler [58.0834973489875]
  Time-Aware Conditional Synthesis (TACS) is a novel approach to conditional generation on diffusion models. It integrates adaptively controlled plug-and-play "online" guidance into a diffusion model, driving samples toward the desired properties.
  arXiv  Detail & Related papers  (2024-11-01T12:59:25Z)
• Generative Diffusion Models for Fast Simulations of Particle Collisions at CERN [3.2686289567336235]
  In High Energy Physics simulations play a crucial role in unraveling the complexities of particle collision experiments within CERN's Large Hadron Collider. Recent advancements highlight the efficacy of diffusion models as state-of-the-art generative machine learning methods. We present the first simulation for Zero Degree Calorimeter (ZDC) at the ALICE experiment based on diffusion models, achieving the highest fidelity compared to existing baselines.
  arXiv  Detail & Related papers  (2024-06-05T13:11:53Z)
• LDMol: A Text-to-Molecule Diffusion Model with Structurally Informative Latent Space Surpasses AR Models [55.5427001668863]
  We present a novel latent diffusion model dubbed LDMol for text-conditioned molecule generation.<n> Experiments show that LDMol outperforms the existing autoregressive baselines on the text-to-molecule generation benchmark.<n>We show that LDMol can be applied to downstream tasks such as molecule-to-text retrieval and text-guided molecule editing.
  arXiv  Detail & Related papers  (2024-05-28T04:59:13Z)
• Diffusion-Driven Generative Framework for Molecular Conformation Prediction [0.66567375919026]
  The rapid advancement of machine learning has revolutionized the precision of predictive modeling in this context. This research introduces a cutting-edge generative framework named method. Method views atoms as discrete entities and excels in guiding the reversal of diffusion.
  arXiv  Detail & Related papers  (2023-12-22T11:49:39Z)
• Prediction of Diblock Copolymer Morphology via Machine Learning [0.0]
  A machine learning approach is presented to accelerate the computation of block polymer morphology evolution over long timescales. In contrast to empirical annihilation models, the proposed approach learns on the continuumally driven defect processes directly from particle-based simulations. This work has implications for directed self-assembly and materials design in micro-electronics, battery materials, and membranes.
  arXiv  Detail & Related papers  (2023-08-31T17:45:34Z)
• Geometric Neural Diffusion Processes [55.891428654434634]
  We extend the framework of diffusion models to incorporate a series of geometric priors in infinite-dimension modelling. We show that with these conditions, the generative functional model admits the same symmetry.
  arXiv  Detail & Related papers  (2023-07-11T16:51:38Z)
• Eliminating Lipschitz Singularities in Diffusion Models [51.806899946775076]
  We show that diffusion models frequently exhibit the infinite Lipschitz near the zero point of timesteps. This poses a threat to the stability and accuracy of the diffusion process, which relies on integral operations. We propose a novel approach, dubbed E-TSDM, which eliminates the Lipschitz of the diffusion model near zero.
  arXiv  Detail & Related papers  (2023-06-20T03:05:28Z)
• Equivariant Parameter Sharing for Porous Crystalline Materials [4.271235935891555]
  Existing methods for crystal property prediction either have constraints that are too restrictive or only incorporate symmetries between unit cells. We develop a model which incorporates the symmetries of the unit cell of a crystal in its architecture and explicitly models the porous structure. Our results confirm that our method performs better than existing methods for crystal property prediction and that the inclusion of symmetry results in a more efficient model.
  arXiv  Detail & Related papers  (2023-04-04T08:33:13Z)
• GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation [102.85440102147267]
  We propose a novel generative model named GeoDiff for molecular conformation prediction. We show that GeoDiff is superior or comparable to existing state-of-the-art approaches.
  arXiv  Detail & Related papers  (2022-03-06T09:47:01Z)
• Normalizing flows for atomic solids [67.70049117614325]
  We present a machine-learning approach, based on normalizing flows, for modelling atomic solids. We report Helmholtz free energy estimates for cubic and hexagonal ice modelled as monatomic water as well as for a truncated and shifted Lennard-Jones system. Our results thus demonstrate that normalizing flows can provide high-quality samples and free energy estimates of solids, without the need for multi-staging or for imposing restrictions on the crystal geometry.
  arXiv  Detail & Related papers  (2021-11-16T18:54:49Z)
• Relativistic particle incoherent scattering by the nuclei of crystal plane atoms [0.0]
  A consistent theory describes the incoherent scattering of classically moving relativistic particles by the nuclei of crystal planes. Basic notions of quantum mechanics are applied to introduce a fundamental compact formula for the mean square incoherent scattering angle per unit length of particle trajectory.
  arXiv  Detail & Related papers  (2021-03-04T16:36:26Z)
• Learning Neural Generative Dynamics for Molecular Conformation Generation [89.03173504444415]
  We study how to generate molecule conformations (textiti.e., 3D structures) from a molecular graph. We propose a novel probabilistic framework to generate valid and diverse conformations given a molecular graph.
  arXiv  Detail & Related papers  (2021-02-20T03:17:58Z)

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.