Explaining deep learning of galaxy morphology with saliency mapping

Bhambra, Prabh; Joachimi, Benjamin; Lahav, O.

doi:10.1093/mnras/stac368

Cited by 13 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3). The principal difference between our approach and the existing ones (see, for example, recent works [56,[84][85][86]) is the usage of 1) the pre-defined training-test split through adversarial validation of the classification accuracy on the inference-like test set, and 2) the specific data augmentation, which allowed us to decrease the difference in galaxy images related to the stellar magnitudes between the GZ2 and inference data sets.…”

Section: Cnn Five-class Morphological Classifiermentioning

confidence: 99%

“…13. Bhambra et al [85] proposed the explainable artificial intelligence (XAI) techniques to measure galactic bar lengths and bulge-to-disk ratio. They used the Hoyle bar length catalog [132] vs. GZ annotated data and demonstrated that XAI works more successfully in predictions of a bar feature.…”

Section: Notes On Problem Points Of Cnn Image-based Galaxy Classifica...mentioning

confidence: 99%

See 1 more Smart Citation

Machine learning technique for morphological classification of galaxies from the SDSS. III. The CNN image-based inference of detailed features

Khramtsov¹,

Vavilova²,

Dobrycheva³

et al. 2022

Kosm. nauka tehnol.

View full text Add to dashboard Cite

This paper follows a series of our works on the applicability of various machine learning methods to morphological galaxy classification (Vavilova et al., 2021, 2022). We exploited the sample of ~315800 low-redshift SDSS DR9 galaxies with absolute stellar magnitudes of −24m < Mr < −19.4m at 0.003 < z < 0.1 redshifts as a target data set for the CNN classifier. Because it is tightly overlapped with the Galaxy Zoo 2 (GZ2) sample, we use these annotated data as the training data set to classify galaxies into 34 detailed features. In the presence of a pronounced difference in visual parameters between galaxies from the GZ2 training data set and galaxies without known morphological parameters, we applied novel procedures, which allowed us for the first time to get rid of this difference for smaller and fainter SDSS galaxies with mr < 17.7. We describe in detail the adversarial validation technique as well as how we managed the optimal train-test split of galaxies from the training data set to verify our CNN model based on the DenseNet-201 realistically. We have also found optimal galaxy image transformations, which help increase the classifier’s generalization ability. We demonstrate for the first time that implication of the CNN model with a train-test split of data sets and size-changing function simulating a decrease in magnitude and size (data augmentation) significantly improves the classification of smaller and fainter SDSS galaxies. It can be considered as another way to improve the human bias for those galaxy images that had a poor vote classification in the GZ project. Such an approach, like autoimmunization, when the CNN classifier, trained on very good galaxy images, is able to retrain bad images from the same homogeneous sample, can be considered co-planar to other methods of combating such a human bias. The most promising result is related to the CNN prediction probability in the classification of detailed features. The accuracy of the CNN classifier is in the range of 83.3—99.4 % depending on 32 features (exception is for “disturbed” (68.55 %) and “arms winding medium” (77.39 %) features). As a result, for the first time, we assigned the detailed morphological classification for more than 140000 low-redshift galaxies, especially at the fainter end. A visual inspection of the samples of galaxies with certain morphological features allowed us to reveal typical problem points of galaxy image classification by shape and features from the astronomical point of view. The morphological catalogs of low-redshift SDSS galaxies with the most interesting features are available through the UkrVO website (http://ukr-vo.org/galaxies/) and VizieR.

show abstract

Section: Cnn Five-class Morphological Classifiermentioning

confidence: 99%

Section: Notes On Problem Points Of Cnn Image-based Galaxy Classifica...mentioning

confidence: 99%

Machine learning technique for morphological classification of galaxies from the SDSS. III. The CNN image-based inference of detailed features

Khramtsov¹,

Vavilova²,

Dobrycheva³

et al. 2022

Kosm. nauka tehnol.

View full text Add to dashboard Cite

show abstract

“…It compares the average relevance maps and the topo-plots for binary masks for various methods, including LRP-based methods. [17] uses saliency mapping by adding Gaussian noise to the input images. Its data set contains many kinds of galaxy images, and it uses some data augmentation techniques such as random rotations and flips.…”

Section: Introductionmentioning

confidence: 99%

“…We call each divided sub-image a part, and then we make each modified image by LIME based methods [7,13] feature (or feature masks) [10,11,15,22,25] LRP-based methods [14,16,21] others [9, 12, 17-20, 23, 24] Table 1 compares each paper from the related works by the category of the method used. Among [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25], [6,8] use the difference of the input values of a neural network to generate explanations. We can compare them with our method.…”

Section: Introductionmentioning

confidence: 99%

An XAI method for convolutional neural networks in self-driving cars

Kim

Joe

2022

PLoS ONE

View full text Add to dashboard Cite

eXplainable Artificial Intelligence (XAI) is a new trend of machine learning. Machine learning models are used to predict or decide something, and they derive output based on a large volume of data set. Here, the problem is that it is hard to know why such prediction was derived, especially when using deep learning models. It makes the models unreliable in the case of reliability-critical applications. So, it is required to explain how they derived such output. It is a reliability-critical application for self-driving cars because the mistakes made by the computers inside them can lead to critical accidents. So, it is necessary to adopt XAI models in this field. In this paper, we propose an XAI method based on computing and explaining the difference of the output values of the neurons in the last hidden layer of convolutional neural networks. First, we input the original image and some modified images of it. Then we derive output values for each image and compare these values. Then, we introduce the Sensitivity Analysis technique to explain which parts of the original image are needed to distinguish the category. In detail, we divide the image into several parts and fill these parts with shades. First, we compute the influence value on the vector indicating the last hidden layer of the model for each of these parts. Then we draw shades whose darkness is in proportion to the influence values. The experimental results show that our approach for XAI in self-driving cars finds the parts needed to distinguish the category of these images accurately.

show abstract

“…Gupta et al (2022) introduced a continuous depth version of the Residual Network called Neural Ordinary Differential Equations (NODE) which obtained an accuracy between 91%-95% depending on the classifications. DL methods used for galaxy morphological classifications are discussed in Tuccillo et al (2016), Khan et al (2019), Ghosh et al (2020), Bhambra et al (2022), Zhang et al (2022) and Vavilova et al (2022).…”

mentioning

confidence: 99%

Unsupervised Galaxy Morphological Visual Representation with Deep Contrastive Learning

Wei¹,

Li²,

Lu³

et al. 2022

PASP

View full text Add to dashboard Cite

Galaxy morphology reflects structural properties that contribute to the understanding of the formation and evolution of galaxies. Deep convolutional networks have proven to be very successful in learning hidden features that allow for unprecedented performance in the morphological classification of galaxies. Such networks mostly follow the supervised learning paradigm, which requires sufficient labeled data for training. However, the labeling of a million galaxies is an expensive and complicated process, particularly for forthcoming survey projects. In this paper, we present an approach, based on contrastive learning, with aim of learning galaxy morphological visual representation using only unlabeled data. Considering the properties of low semantic information and contour dominated of galaxy images, the feature extraction layer of the proposed method incorporates vision transformers and a convolutional network to provide rich semantic representation via the fusion of multi-hierarchy features. We train and test our method on three classifications of data sets from Galaxy Zoo 2 and SDSS-DR17, and four classifications from Galaxy Zoo DECaLS. The testing accuracy achieves 94.7%, 96.5% and 89.9%, respectively. The experiment of cross validation demonstrates our model possesses transfer and generalization ability when applied to new data sets. The code that reveals our proposed method and pretrained models are publicly available and can be easily adapted to new surveys. 6 6 https://github.com/kustcn/galaxy_contrastive

show abstract

Explaining deep learning of galaxy morphology with saliency mapping

Cited by 13 publications

References 28 publications

Machine learning technique for morphological classification of galaxies from the SDSS. III. The CNN image-based inference of detailed features

Machine learning technique for morphological classification of galaxies from the SDSS. III. The CNN image-based inference of detailed features

An XAI method for convolutional neural networks in self-driving cars

Unsupervised Galaxy Morphological Visual Representation with Deep Contrastive Learning

Contact Info

Product

Resources

About