arXiv subjects

Yan Cen

Publications and source records attributed to Yan Cen.

4 recordsLinked to original sources

arXiv2024

VQCrystal: Leveraging Vector Quantization for Discovery of Stable Crystal Structures

Discovering functional crystalline materials through computational methods remains a formidable challenge in materials science. Here, we introduce VQCrystal, an innovative deep learning framework that leverages discrete latent representations to overcome key limitations in current approaches to crystal generation and inverse design. VQCrystal employs a hierarchical VQ-VAE architecture to encode global and atom-level crystal features, coupled with a machine learning-based inter-atomic potential(IAP) model and a genetic algorithm to realize property-targeted inverse design. Benchmark evaluations on diverse datasets demonstrate VQCrystal's advanced capabilities in representation learning and novel crystal discovery. Notably, VQCrystal achieves state-of-the-art performance with 91.93\% force validity and a Fr\'echet Distance of 0.152 on MP-20, indicating both strong validity and high diversity in the sampling process. To demonstrate real-world applicability, we apply VQCrystal for both 3D and 2D material design. For 3D materials, the density-functional theory validation confirmed that 63.04\% of bandgaps and 99\% of formation energies of the 56 filtered materials matched the target range. Moreover, 437 generated materials were validated as existing entries in the full database outside the training set. For the discovery of 2D materials, 73.91\% of 23 filtered structures exhibited high stability with formation energies below -1 eV/atom. Our results highlight VQCrystal's potential to accelerate the discovery of novel materials with tailored properties.

cond-mat.mtrl-sci↗

arXiv2024

CTGNN: Crystal Transformer Graph Neural Network for Crystal Material Property Prediction

The combination of deep learning algorithm and materials science has made significant progress in predicting novel materials and understanding various behaviours of materials. Here, we introduced a new model called as the Crystal Transformer Graph Neural Network (CTGNN), which combines the advantages of Transformer model and graph neural networks to address the complexity of structure-properties relation of material data. Compared to the state-of-the-art models, CTGNN incorporates the graph network structure for capturing local atomic interactions and the dual-Transformer structures to model intra-crystal and inter-atomic relationships comprehensively. The benchmark carried on by the proposed CTGNN indicates that CTGNN significantly outperforms existing models like CGCNN and MEGNET in the prediction of formation energy and bandgap properties. Our work highlights the potential of CTGNN to enhance the performance of properties prediction and accelerates the discovery of new materials, particularly for perovskite materials.

cond-mat.mtrl-sci↗

arXiv2023

Using Large Language Model to Solve and Explain Physics Word Problems Approaching Human Level

Our work demonstrates that large language model (LLM) pre-trained on texts can not only solve pure math word problems, but also physics word problems, whose solution requires calculation and inference based on prior physical knowledge. We collect and annotate the first physics word problem dataset-PhysQA, which contains over 1000 junior high school physics word problems (covering Kinematics, Mass&Density, Mechanics, Heat, Electricity). Then we use OpenAI' s GPT3.5 to generate the answer of these problems and found that GPT3.5 could automatically solve 49.3% of the problems through zero-shot learning and 73.2% through few-shot learning. This result demonstrates that by using similar problems and their answers as prompt, LLM could solve elementary physics word problems approaching human level performance. In addition to solving problems, GPT3.5 can also summarize the knowledge or topics covered by the problems, provide relevant explanations, and generate new physics word problems based on the input. Our work is the first research to focus on the automatic solving, explanation, and generation of physics word problems across various types and scenarios, and we achieve an acceptable and state-of-the-art accuracy. This underscores the potential of LLMs for further applications in secondary education.

cs.CL↗

arXiv2023

Discovery of Stable Hybrid Organic-inorganic Double Perovskites for High-performance Solar Cells via Machine-learning Algorithms and Crystal Graph Convolution Neural Network Method

Hybrid peroskite solar cells are newly emergent high-performance photovoltaic devices, which suffer from disadvantages such as toxic elements, short-term stabilities, and so on. Searching for alternative perovskites with high photovoltaic performances and thermally stabilities is urgent in this field. In this work, stimulated by the recently proposed materials-genome initiative project, firstly we build classical machine-learning algorithms for the models of formation energies, bangdaps and Deybe temperatures for hybrid organic-inorganic double perovskites, then we choose the high-precision models to screen a large scale of double-perovskite chemical space, to filter out good pervoskite candidates for solar cells. We also analyze features of importances for the the three target properties to reveal the underlying mechanisms and discover the typical characteristics of high-performances double perovskites. Secondly we adopt the Crystal graph convolution neural network (CGCNN), to build precise model for bandgaps of perovskites for further filtering. Finally we use the ab-initio method to verify the results predicted by the CGCNN method, and find that, six out of twenty randomly chosen (CH3)2NH2-based HOIDP candidates possess finite bandgaps, and especially, (CH3)2NH2AuSbCl6 and (CH3)2NH2CsPdF6 possess the bandgaps of 0.633 eV and 0.504 eV, which are appropriate for photovoltaic applications. Our work not only provides a large scale of potential high-performance double-perovskite candidates for futural experimental or theoretical verification, but also showcases the effective and powerful prediction of the combined ML and CGCNN method proposed for the first time here.

cond-mat.mtrl-sci↗