Efficient few-shot machine learning for classification of EBSD patterns.
Efficient few-shot machine learning for classification of EBSD patterns.
复制标题
DOI：
10.1038/s41598-021-87557-5
复制
发表时间：
2021-04-14
影响因子：
4.6
通讯作者：
Vecchio KS
中科院分区：
文献类型：
--
作者：
Kaufmann K;Lane H;Liu X;Vecchio KS
深度学习正迅速成为解决一系列材料科学问题的标准方法，特别是在计算机视觉领域。然而，大到足以从头开始训练神经网络的标记数据集可能很难收集。加速深度学习模型（如卷积神经网络）训练的一种方法是从不相关图像分类问题上训练的模型中转移权重，通常称为迁移学习。以前学到的强大的特征提取器可以在不影响性能的情况下对新的分类问题进行微调。迁移学习还可以提高使用少量数据训练模型的结果，称为少射学习。在此，我们测试了少量迁移学习方法的有效性，用于将电子背散射衍射（EBSD）图案图像分类到点群中的六个空间群。将训练历史和性能指标与从头开始训练的相同架构的模型进行比较。为了使这种方法更易于解释，使用了过滤器、激活图和Shapley值的可视化来提供对模型操作的洞察。应用双相材料对现实世界的相识别和区分进行了验证，而传统方法很难对双相材料进行分析。
Deep learning is quickly becoming a standard approach to solving a range of materials science objectives, particularly in the field of computer vision. However, labeled datasets large enough to train neural networks from scratch can be challenging to collect. One approach to accelerating the training of deep learning models such as convolutional neural networks is the transfer of weights from models trained on unrelated image classification problems, commonly referred to as transfer learning. The powerful feature extractors learned previously can potentially be fine-tuned for a new classification problem without hindering performance. Transfer learning can also improve the results of training a model using a small amount of data, known as few-shot learning. Herein, we test the effectiveness of a few-shot transfer learning approach for the classification of electron backscatter diffraction (EBSD) pattern images to six space groups within the point group. Training history and performance metrics are compared with a model of the same architecture trained from scratch. In an effort to make this approach more explainable, visualization of filters, activation maps, and Shapley values are utilized to provide insight into the model’s operations. The applicability to real-world phase identification and differentiation is demonstrated using dual phase materials that are challenging to analyze with traditional methods.
登录
查看更多内容
影响因子：
2.2
作者：
Foden, A.;Collins, D. M.;Britton, T. B.
通讯作者：
Britton, T. B.
影响因子：
3.3
作者：
DeCost, Brian L.;Holm, Elizabeth A.
通讯作者：
Holm, Elizabeth A.
影响因子：
3.9
作者：
Adadi, Amina;Berrada, Mohammed
通讯作者：
Berrada, Mohammed
影响因子：
9.4
作者：
Ding, Z.;Pascal, E.;De Graef, M.
通讯作者：
De Graef, M.
影响因子：
2.2
作者：
Hielscher, Ralf;Bartel, Felix;Britton, Thomas Benjamin
通讯作者：
Britton, Thomas Benjamin