Exploring TeV Candidates of Fermi Blazars through Machine Learning

Abstract

In this work, we make use of a supervised machine-learning algorithm based on Logistic Regression (LR) to select TeV blazar candidates from the 4FGL-DR2/4LAC-DR2, 3FHL, 3HSP, and 2BIGB catalogs. LR constructs a hyperplane based on a selection of optimal parameters, named features, and hyperparameters whose values control the learning process and determine the values of features that a learning algorithm ends up learning, to discriminate TeV blazars from non-TeV blazars. In addition, it gives the probability (or logistic) that a source may be considered a TeV blazar candidate. Non-TeV blazars with logistics greater than 80% are considered high-confidence TeV candidates. Using this technique, we identify 40 high-confidence TeV candidates from the 4FGL-DR2/4LAC-DR2 blazars and we build the feature hyperplane to distinguish TeV and non-TeV blazars. We also calculate the hyperplanes for the 3FHL, 3HSP, and 2BIGB. Finally, we construct the broadband spectral energy distributions for the 40 candidates, testing for their detectability with various instruments. We find that seven of them are likely to be detected by existing or upcoming IACT observatories, while one could be observed with extensive air shower particle detector arrays.

A Study of Broad Emission Line and Doppler Factor Estimation for Fermi Blazars

“…In order to effectively separate these two classes, we employed a kind of machine-learning (ML) method to establish a dividing line. Recently, ML methods, such as support vector machine (SVM), artificial neural networks, K-nearest neighbors, etc., have been widely used in astronomy; see Kang et al (2019), Kovačević et al (2019), Xu et al (2020), Zhu et al (2021), Xiao et al (2022bXiao et al ( , 2022cXiao et al ( , 2023, and Zhu et al (2023). The SVM model can easily handle both linear and nonlinear classification problems by choosing different kernel functions.…”

Characterizing the Emission Region Properties of Blazars

“…In this case, we have 754 FSRQs using the model of external soft photons coming from the BLR; this is consistent with the assumption of a soft photon origin for FSRQs in the literature (e.g., Tan et al 2020) and also consistent with the facts that the FSRQs show significant broad emission lines and the emission lines contribute to the EC component significantly (Xiao et al 2022a). The rest of the five FSRQs are considered as HSPs, and the HSPs are naturally considered as TeV candidates (Zhu et al 2023). The TeV emission could be severely absorbed by interacting with BLR soft photons, thus we assume these five FSRQs with soft photons are from the DT.…”

Classification of Fermi BCUs Using Machine Learning

“…SVM (Vapnik 2013;Kang et al 2019a;Zhu et al 2021b) is used to solve the separating hyperplane that correctly partitions the training data set and has the largest geometric separation. ANN (Bishop1995; Kovačević et al 2019;Zhu et al 2023) is used, which simulates the working principle of nerve cells, and composes an artificial neuron into a network structure to establish a relationship between input and output. We use KNN (Arthur & Vassilvitskii 2006;Xu et al 2020), which determines the category of the sample to be divided according to the category of the nearest one or several samples.…”