Machine-learning accelerated identification of exfoliable two-dimensional materials

• URL: http://arxiv.org/abs/2207.12118v1
• Date: Mon, 18 Jul 2022 14:48:53 GMT
• Title: Machine-learning accelerated identification of exfoliable two-dimensional materials
• Authors: Mohammad Tohidi Vahdat, Kumar Agrawal Varoon, and Giovanni Pizzi
• Abstract summary: Two-dimensional (2D) materials have been a central focus of recent research because they host a variety of properties. It is crucial to be able to identify accurately and efficiently if bulk three-dimensional (3D) materials are formed by layers held together by a weak binding energy. We develop a machine-learning (ML) approach that, combined with a fast preliminary geometrical screening, is able to efficiently identify potentially exfoliable materials.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Two-dimensional (2D) materials have been a central focus of recent research because they host a variety of properties, making them attractive both for fundamental science and for applications. It is thus crucial to be able to identify accurately and efficiently if bulk three-dimensional (3D) materials are formed by layers held together by a weak binding energy that, thus, can be potentially exfoliated into 2D materials. In this work, we develop a machine-learning (ML) approach that, combined with a fast preliminary geometrical screening, is able to efficiently identify potentially exfoliable materials. Starting from a combination of descriptors for crystal structures, we work out a subset of them that are crucial for accurate predictions. Our final ML model, based on a random forest classifier, has a very high recall of 98\%. Using a SHapely Additive exPlanations (SHAP) analysis, we also provide an intuitive explanation of the five most important variables of the model. Finally, we compare the performance of our best ML model with a deep neural network architecture using the same descriptors. To make our algorithms and models easily accessible, we publish an online tool on the Materials Cloud portal that only requires a bulk 3D crystal structure as input. Our tool thus provides a practical yet straightforward approach to assess whether any 3D compound can be exfoliated into 2D layers.

Related papers

• MaterialSeg3D: Segmenting Dense Materials from 2D Priors for 3D Assets [63.284244910964475]
  We propose a 3D asset material generation framework to infer underlying material from the 2D semantic prior. Based on such a prior model, we devise a mechanism to parse material in 3D space.
  arXiv  Detail & Related papers  (2024-04-22T07:00:17Z)
• Leveraging Large-Scale Pretrained Vision Foundation Models for Label-Efficient 3D Point Cloud Segmentation [67.07112533415116]
  We present a novel framework that adapts various foundational models for the 3D point cloud segmentation task. Our approach involves making initial predictions of 2D semantic masks using different large vision models. To generate robust 3D semantic pseudo labels, we introduce a semantic label fusion strategy that effectively combines all the results via voting.
  arXiv  Detail & Related papers  (2023-11-03T15:41:15Z)
• EGFN: Efficient Geometry Feature Network for Fast Stereo 3D Object Detection [51.52496693690059]
  Fast stereo based 3D object detectors lag far behind high-precision oriented methods in accuracy. We argue that the main reason is the missing or poor 3D geometry feature representation in fast stereo based methods. The proposed EGFN outperforms YOLOStsereo3D, the advanced fast method, by 5.16% on mAP$_3d$ at the cost of merely additional 12 ms.
  arXiv  Detail & Related papers  (2021-11-28T05:25:36Z)
• How to See Hidden Patterns in Metamaterials with Interpretable Machine Learning [82.67551367327634]
  We develop a new interpretable, multi-resolution machine learning framework for finding patterns in the unit-cells of materials. Specifically, we propose two new interpretable representations of metamaterials, called shape-frequency features and unit-cell templates.
  arXiv  Detail & Related papers  (2021-11-10T21:19:02Z)
• Translational Symmetry-Aware Facade Parsing for 3D Building Reconstruction [11.263458202880038]
  In this paper, we present a novel translational symmetry-based approach to improving the deep neural networks. We propose a novel scheme to fuse anchor-free detection in a single stage network, which enables the efficient training and better convergence. We employ an off-the-shelf rendering engine like Blender to reconstruct the realistic high-quality 3D models using procedural modeling.
  arXiv  Detail & Related papers  (2021-06-02T03:10:51Z)
• Learning Feature Aggregation for Deep 3D Morphable Models [57.1266963015401]
  We propose an attention based module to learn mapping matrices for better feature aggregation across hierarchical levels. Our experiments show that through the end-to-end training of the mapping matrices, we achieve state-of-the-art results on a variety of 3D shape datasets.
  arXiv  Detail & Related papers  (2021-05-05T16:41:00Z)
• Predicting Material Properties Using a 3D Graph Neural Network with Invariant Local Descriptors [0.4956709222278243]
  Accurately predicting material properties is critical for discovering and designing novel materials. Among the machine learning methods, graph convolution neural networks (GCNNs) have been one of the most successful ones. We propose an adaptive GCNN with novel convolutions that model interactions among all neighboring atoms in three-dimensional space simultaneously.
  arXiv  Detail & Related papers  (2021-02-16T19:56:54Z)
• Generating 3D structures from a 2D slice with GAN-based dimensionality expansion [0.0]
  Generative adversarial networks (GANs) can be trained to generate 3D image data, which is useful for design optimisation. We introduce a generative adversarial network architecture, SliceGAN, which is able to synthesise high fidelity 3D datasets using a single representative 2D image.
  arXiv  Detail & Related papers  (2021-02-10T18:46:17Z)
• Generative VoxelNet: Learning Energy-Based Models for 3D Shape Synthesis and Analysis [143.22192229456306]
  This paper proposes a deep 3D energy-based model to represent volumetric shapes. The benefits of the proposed model are six-fold. Experiments demonstrate that the proposed model can generate high-quality 3D shape patterns.
  arXiv  Detail & Related papers  (2020-12-25T06:09:36Z)
• Molecular machine learning with conformer ensembles [0.0]
  We introduce multiple deep learning models that expand upon key architectures such as ChemProp and Schnet. We then benchmark the performance trade-offs of these models on 2D, 3D and 4D representations in the prediction of drug activity. The new architectures perform significantly better than 2D models, but their performance is often just as strong with a single conformer as with many.
  arXiv  Detail & Related papers  (2020-12-15T17:44:48Z)
• 3DMaterialGAN: Learning 3D Shape Representation from Latent Space for Materials Science Applications [7.449993399792031]
  3DMaterialGAN is capable of recognizing and synthesizing individual grains whose morphology conforms to a given 3D polycrystalline material microstructure. We show that this method performs comparably or better than state-of-the-art on benchmark annotated 3D datasets. This framework lays the foundation for the recognition and synthesis of polycrystalline material microstructures.
  arXiv  Detail & Related papers  (2020-07-27T21:55:16Z)

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.