Related papers: Interpreting automatic AGN classifiers with salien…
Generative Adversarial Networks (GANs) have recently achieved impressive results for many real-world applications, and many GAN variants have emerged with improvements in sample quality and training stability. However, they have not been…
We aim to study the effect of environment on the presence and fuelling of Active Galactic Nuclei (AGN) in massive galaxy clusters. We explore the use of different AGN detection techniques with the goal of selecting AGN across a broad range…
We use machine learning techniques to investigate their performance in classifying active galactic nuclei (AGNs), including X-ray selected AGNs (XAGNs), infrared selected AGNs (IRAGNs), and radio selected AGNs (RAGNs). Using known physical…
The black-box nature of the deep networks makes the explanation for "why" they make certain predictions extremely challenging. Saliency maps are one of the most widely-used local explanation tools to alleviate this problem. One of the…
In the last few decades, significant achievements have been attained in predicting where humans look at images through different computational models. However, how to determine contributions of different visual features to overall saliency…
How can we improve the facial soft-biometric classification with help of the human visual system? This paper explores the use of saliency which is equivalent to the human visual system to classify Age, Gender and Facial Expression…
Active galactic nuclei (AGN) are typically identified through radio, mid-infrared, or X-ray emission or through the presence of broad and/or narrow emission lines. AGN can also leave an imprint on a galaxy's spectral energy distribution…
Graph Neural Networks (GNNs) have emerged as the predominant approach for learning over graph-structured data. However, most GNNs operate as black-box models and require post-hoc explanations, which may not suffice in high-stakes scenarios…
Active Galactic Nuclei (AGNs) are characterized by emission of radiation over more than 10 orders of magnitude in frequency. Therefore, the execution of extensive surveys of the sky, with different types of detectors, is providing the…
Convolutional neural networks (CNNs) offer great machine learning performance over a range of applications, but their operation is hard to interpret, even for experts. Various explanation algorithms have been proposed to address this issue,…
We propose a new method for identifying active galactic nuclei (AGN) in low mass ($\rm M_*\leq10^{10}M_\odot$) galaxies. This method relies on spectral energy distribution (SED) fitting to identify galaxies whose radio flux density has an…
Clinical applicability of automated decision support systems depends on a robust, well-understood classification interpretation. Artificial neural networks while achieving class-leading scores fall short in this regard. Therefore, numerous…
In many practical applications, deep neural networks have been typically deployed to operate as a black box predictor. Despite the high amount of work on interpretability and high demand on the reliability of these systems, they typically…
During a geosteering operation the well path is intentionally adjusted in response to the new data acquired while drilling. To achieve consistent high-quality decisions, especially when drilling in complex environments, decision support…
Machine learning (ML) based models have greatly enhanced the traditional materials discovery and design pipeline. Specifically, in recent years, surrogate ML models for material property prediction have demonstrated success in predicting…
With the advancement of technology, machine learning-based analytical methods have pervaded nearly every discipline in modern studies. Particularly, a number of methods have been employed to estimate the redshift of gamma-ray loud active…
As wide-field optical surveys such as Vera Rubin Observatory's Legacy Survey of Space and Time (LSST) begin operations, time-domain astronomy is facing a data revolution, paving the road for new, expanded variability studies. This work…
Supervised artificial neural networks are used to predict useful properties of galaxies in the Sloan Digital Sky Survey, in this instance morphological classifications, spectral types and redshifts. By giving the trained networks unseen…
Methods based on class activation maps (CAM) provide a simple mechanism to interpret predictions of convolutional neural networks by using linear combinations of feature maps as saliency maps. By contrast, masking-based methods optimize a…
Machine learning (ML) tools such as encoder-decoder convolutional neural networks (CNN) can represent incredibly complex nonlinear functions which map between combinations of images and scalars. For example, CNNs can be used to map…