Comparison of Multi-class and Binary Classification Machine Learning Models in Identifying Strong Gravitational Lenses

Teimoorinia, Hossen; Toyonaga, Robert D.; Fabbro, S.; Bottrell, Connor

doi:10.1088/1538-3873/ab747b

Cited by 13 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A possibility would be to use CNNs optimized for image outlier detection (e.g., Margalef-Bentabol et al 2020), either to identify strong lenses as the outlier class, or simply to exclude cutouts with partial coverage, background artifacts or other anomalies before classification. In addition, although Teimoorinia et al (2020) found little difference with single-band HST images, implementing additional classes for the usual interlopers (spirals, ring galaxies, . .…”

Section: Discussionmentioning

confidence: 89%

Holismokes

Cañameras

Schuldt

Suyu

et al. 2020

A&A

View full text Add to dashboard Cite

We present a systematic search for wide-separation (with Einstein radius θE ≳ 1.5″), galaxy-scale strong lenses in the 30 000 deg2 of the Pan-STARRS 3π survey on the Northern sky. With long time delays of a few days to weeks, these types of systems are particularly well-suited for catching strongly lensed supernovae with spatially-resolved multiple images and offer new insights on early-phase supernova spectroscopy and cosmography. We produced a set of realistic simulations by painting lensed COSMOS sources on Pan-STARRS image cutouts of lens luminous red galaxies (LRGs) with redshift and velocity dispersion known from the sloan digital sky survey (SDSS). First, we computed the photometry of mock lenses in gri bands and applied a simple catalog-level neural network to identify a sample of 1 050 207 galaxies with similar colors and magnitudes as the mocks. Second, we trained a convolutional neural network (CNN) on Pan-STARRS gri image cutouts to classify this sample and obtain sets of 105 760 and 12 382 lens candidates with scores of pCNN > 0.5 and > 0.9, respectively. Extensive tests showed that CNN performances rely heavily on the design of lens simulations and the choice of negative examples for training, but little on the network architecture. The CNN correctly classified 14 out of 16 test lenses, which are previously confirmed lens systems above the detection limit of Pan-STARRS. Finally, we visually inspected all galaxies with pCNN > 0.9 to assemble a final set of 330 high-quality newly-discovered lens candidates while recovering 23 published systems. For a subset, SDSS spectroscopy on the lens central regions proves that our method correctly identifies lens LRGs at z ∼ 0.1–0.7. Five spectra also show robust signatures of high-redshift background sources, and Pan-STARRS imaging confirms one of them as a quadruply-imaged red source at zs = 1.185, which is likely a recently quenched galaxy strongly lensed by a foreground LRG at zd = 0.3155. In the future, high-resolution imaging and spectroscopic follow-up will be required to validate Pan-STARRS lens candidates and derive strong lensing models. We also expect that the efficient and automated two-step classification method presented in this paper will be applicable to the ∼4 mag deeper gri stacks from the Rubin Observatory Legacy Survey of Space and Time (LSST) with minor adjustments.

show abstract

Section: Discussionmentioning

confidence: 89%

Holismokes

Cañameras

Schuldt

Suyu

et al. 2020

A&A

View full text Add to dashboard Cite

show abstract

“…The F1 value is usually utilized to evaluate the comprehensive performance of the binary classification model, which simultaneously considers classification precision and the recall rate. Equations – are used to calculate the F1 value pr normale normalc normali normals normali normalo normaln = normalT normalP normalT normalP + normalF normalP re normalc normala normall normall = normalT normalP normalT normalP + normalF normalN normalF 1 = 2 ∗ pr normale normalc normali normals normali normalo normaln · normalr normale normalc normala normall normall pr normale normalc normali normals normali normalo normaln + normalr normale normalc normala normall normall where true positive (TP) represents the sample numbers correctly predicted as a positive example, false positive (FP) represents the sample numbers that are incorrectly predicted as positive, and false negative (FN) represents the sample numbers that are incorrectly predicted as a counter example.…”

Section: Resultsmentioning

confidence: 99%

Multiscale Turbulence Characteristics of Gas–Liquid Pressure Fluctuations in an Industrial Quatro Static Mixer

Yu,

Li,

Meng

et al. 2024

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

The gas−liquid two-phase flow is of great significance to industrial production. The nonlinear turbulence characteristics of pressure fluctuation in a Quatro static mixer (QSM) are experimentally investigated under different superficial gas/liquid velocities to further study the flow characteristics of the gas−liquid two-phase. It is shown that there is a generalized self-similarity scaling law in the QSM, which is obtained by fitting the structure function of pressure fluctuation signals at different inlet and outlet flow velocities. Through self-similarity scaling law analysis, the scaling exponent parameters of the structure function of two-phase flow pressure fluctuation under apparent liquid velocity (U L ) = 0.028 m/s and different apparent gas velocities (U G ) in QSM are calculated. The calculation results show that the singular scaling exponent (γ) and intermittency parameter (β) are in the ranges of −0.142−0.311 and 0.286−0.966, respectively. This indicates that there exists an extended self-similarity scaling law in QSM under a low Reynolds number (Re). The R/S and autocorrelation analysis reveal that the detailed signals at the inlet and outlet have three similar flow behaviors, which consist of the liquid turbulence at the inlet, the bubble movement, breakup and coalescence, and the liquid swirling flow. The nonlinear analysis shows that flow in QSM presents a pseudohomogeneous flow pattern when the Hurst index (H) = 0.4−0.6, while it presents a heterogeneous flow pattern when H = 0.6−0.72. Support vector machine (SVM) and Random Forest models (RF) are utilized to better predict the flow pattern, with the recognition precision reaching 99.08 and 99.39%, respectively.

show abstract

“…Another field of application of AI/ML in cosmological analysis, as indicated in this cluster by the term "gravitational lensing," is the study of gravitational lenses. In this sense, beyond the discoveries produced by AI/ML in this field (Mirzoyan et al, 2019;Ostrovski et al, 2017;Teimoorinia et al, 2020), it should be mentioned that one of the obstacles in applying DL in the search for this peculiar astrophysical phenomenon is the scarcity of data with which to carry out the training phase. For this reason, simulated lenses are often used, usually on a galactic scale.…”

Section: Study Of Cosmology Using Ai and ML Methods (Yellow Cluster)mentioning

confidence: 99%

On the application of machine learning in astronomy and astrophysics: A text‐mining‐based scientometric analysis

Rodríguez

Rodríguez‐Rodríguez

Woo

2022

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Since the beginning of the 21st century, the fields of astronomy and astrophysics have experienced significant growth at observational and computational levels, leading to the acquisition of increasingly huge volumes of data. In order to process this vast quantity of information, artificial intelligence (AI) techniques are being combined with data mining to detect patterns with the aim of modeling, classifying or predicting the behavior of certain astronomical phenomena or objects. Parallel to the exponential development of the aforementioned techniques, the scientific output related to the application of AI and machine learning (ML) in astronomy and astrophysics has also experienced considerable growth in recent years. Therefore, the increasingly abundant articles make it difficult to monitor this field in terms of which research topics are the most prolific or novel, or which countries or authors are leading them. In this article, a text‐mining‐based scientometric analysis of scientific documents published over the last three decades on the application of AI and ML in the fields of astronomy and astrophysics is presented. The VOSviewer software and data from the Web of Science (WoS) are used to elucidate the evolution of publications in this research field, their distribution by country (including co‐authorship), the most relevant topics addressed, and the most cited elements and most significant co‐citations according to publication source and authorship. The obtained results demonstrate how application of AI/ML to the fields of astronomy/astrophysics represents an established and rapidly growing field of research that is crucial to obtaining scientific understanding of the universe. This article is categorized under: Algorithmic Development > Text Mining Technologies > Machine Learning Application Areas > Science and Technology

show abstract

Comparison of Multi-class and Binary Classification Machine Learning Models in Identifying Strong Gravitational Lenses

Cited by 13 publications

References 22 publications

Holismokes

Holismokes

Multiscale Turbulence Characteristics of Gas–Liquid Pressure Fluctuations in an Industrial Quatro Static Mixer

On the application of machine learning in astronomy and astrophysics: A text‐mining‐based scientometric analysis

Contact Info

Product

Resources

About