← Back to all news
arXiv Machine Learning By Joseph Agada, Yishu Wang, Arpan Biswas

CrystalMO-TuRBO: Multi-Objective Trust-Region Bayesian Optimization for High-precision Joint Crystal Structure Refinement

Read the original on arXiv Machine Learning →

CrystalMO‑TuRBO is a multi‑objective trust‑region Bayesian optimization framework designed for joint crystal structure refinement using X‑ray and neutron diffraction data. It treats the discrepancies from each modality as separate objectives, first exploring the parameter space globally with parallel Bayesian optimization across multiple scalarizations, then refining locally within a shrinking trust region to achieve high‑precision solutions. Experiments on single‑crystal Ho₂Ti₂O₇ data show that this two‑phase approach improves convergence, robustness, and parameter precision over traditional least‑squares, likelihood‑based, and single‑objective Bayesian methods.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv AI
1d ago

XDecomposer: Learning Prior-Free Set Decomposition for Multiphase X-ray Diffraction

XDecomposer is a prior‑free framework that jointly decomposes and identifies multiphase X‑ray diffraction patterns without requiring candidate phase lists, structural templates, or knowledge of the number of phases. It treats multiphase analysis as a set prediction problem, inferring an unordered set of phase‑resolved components, their mixture proportions, and structural representations within a single architecture. Experiments on simulated and experimental data demonstrate improved reconstruction accuracy and phase identification across diverse chemical systems, with strong generalization to unseen mixtures.

By Hanyu Gao, Bin Cao, Yunyue Su, Tong-Yi Zhang, Qiang Liu
More like this →
arXiv Machine Learning
Sep 3

Scalable Bayesian Optimization of Composite Functions for Image-Based Inverse Problems in Materials Characterization

The paper introduces Scalable Bayesian Optimization of Composite Functions (SBOCF) for efficiently estimating physical parameters from scientific images, specifically targeting electron microscopy PACBED patterns. SBOCF leverages the composite structure of the image-matching objective, reducing modeled outputs from 24,649 to 11 by using patch-level summaries and correction terms. With only 50 simulator evaluations, SBOCF outperformed standard Bayesian optimization, achieving up to 290× lower median SSE on synthetic SrTiO3 benchmarks and producing accurate parameter estimates on experimental data without task-specific pretraining.

By Dasol Yoon, Poompol Buathong, Chia-Hao Lee, Yujia Zhang, David A. Muller, Peter I. Frazier
benchmarks
More like this →
arXiv Machine Learning
6d ago

Riemannian Flow Models with Reinforcement Learning for Molecular Crystal Structure Prediction

The paper introduces CG-OMatG, an equivariant Riemannian flow-based generative model that predicts molecular crystal structures using a coarse-grained, hierarchical representation. It treats molecules as rigid bodies, performs inter- and intra-molecular message passing, and learns to reconstruct molecule centroids, orientations, and lattice parameters conditioned on chemical species and conformer geometry. The model is trained on OMC25 and CSD datasets, fine-tuned with policy gradient reinforcement learning to favor low-energy structures, and validated against a CSP blind test benchmark using COMPACK packing-similarity analysis.

By Thomas Egg, Harry Winston Sullivan, Maya M. Martirossyan, Philipp H\"ollmer, Cheng Zeng, Adrian Roitberg, Mingjie Liu, Richard Hennig, Sapna Sarupria, Ellad B. Tadmor, Stefano Martiniani
reinforcement-learningfine-tuningbenchmarks
More like this →