Attention-gating for improved radio galaxy classification
Attention-gating for improved radio galaxy classification
复制标题
DOI：
10.1093/mnras/staa3946
复制
发表时间：
2020-12
期刊：
影响因子：
--
通讯作者：
Micah Bowles;A. Scaife;Fiona A. M. Porter;Hongming Tang;David J. Bastien
中科院分区：
文献类型：
--
作者：
Micah Bowles;A. Scaife;Fiona A. M. Porter;Hongming Tang;David J. Bastien
在这项工作中，我们引入注意力作为一种最先进的机制，使用卷积神经网络对射电星系进行分类。我们提出了一个基于注意力的模型，其性能与以前的分类器相当，同时使用的参数比该领域下一个最小的经典CNN应用程序少50%以上。我们定量地展示了注意力门控中使用的归一化和聚合方法的选择如何影响单个模型的输出，并表明所得到的注意力图可以用来解释模型所做的分类选择。我们观察到，由我们的模型识别的显著区域与专家人类分类器将参加进行等效分类的区域非常一致。我们表明，虽然选择的归一化和聚合可能只会最小限度地影响个别模型的性能，它可以显着影响各自的注意力地图的可解释性，并通过选择一个模型，以及天文学家如何分类无线电源的眼睛，用户可以采用更有效的方式模型。
In this work, we introduce attention as a state-of-the-art mechanism for classification of radio galaxies, using convolutional neural networks. We present an attention-based model that performs on par with previous classifiers while using more than 50 per cent fewer parameters than the next smallest classic CNN application in this field. We demonstrate quantitatively how the selection of normalization and aggregation methods used in attention-gating can affect the output of individual models, and show that the resulting attention maps can be used to interpret the classification choices made by the model. We observe that the salient regions identified by the our model align well with the regions an expert human classifier would attend to make equivalent classifications. We show that while the selection of normalization and aggregation may only minimally affect the performance of individual models, it can significantly affect the interpretability of the respective attention maps and by selecting a model which aligns well with how astronomers classify radio sources by eye, a user can employ the model in a more effective manner.