44 research outputs found
Comparing Optical Variability of Type 1 and Type 2 AGN from the BAT 9 Month Sample Using ASAS-SN and TESS Surveys
No full textWe present an optical variability analysis and comparison of the samples of Seyfert 1 (Sy1) and Seyfert 2 (Sy2) galaxies, selected from the Swift 9 month BAT catalog, using the light curves from Transiting Exoplanet Survey Satellite (TESS) and All-Sky Automated Survey for SuperNovae (ASAS-SN). We measured the normalized excess variance of TESS and ASAS-SN light curves for each target and performed a Kolmogorov–Smirnov test between the two samples, where our results showed significant differences. This is consistent with predictions from the unification model, where Seyfert 2s are obscured by the larger scale dust torus and their variability is suppressed. This variability difference is independent of the luminosity, Eddington ratio, or black hole mass, further supporting geometrical unification models. We searched the dependence of the normalized excess variance of Sy1s on absolute magnitudes, Eddington ratio, and black hole mass, where our results are consistent with relations found in the literature. Finally, a small subsample of changing-look (CL) active galactic nuclei (AGNs) that transitioned during the time frame of the ASAS-SN light curves, with their variability amplitudes changing according to the classification, have larger variability as type 1s and smaller as 2s. The change of variability amplitudes can be used to better pinpoint when the type transition occurred. The consistency trend of the variability amplitude differences between Sy1s and Sy2s and between CL AGNs in 1 or 2 stages suggests that variability can be a key factor in shedding light on the CL AGN or the dichotomy between Sy1 or Sy2 populations
Deep learning with limited data
No full textThe field of deep learning has made significant strides in recent years, largely due to the rapid increase in computational power and the abundance of data. With the help of neural network architectures, deep learning has made significant advancements in fields such as visual image analysis and natural language processing, by autonomously identifying pat- terns in raw data through learning from large-scale datasets. One reason for deep learning’s success is its over-parameterization, which allows it to be optimized to global optimality using local search heuristics. And there is a growing trend in designing larger neural net- work models with bigger data sets. Unfortunately, the cost of training these models scales with the product of the number of parameters and the number of data points, which can quickly become a burden. Additionally, big data with annotations is not always readily available.To address this, our research aims to tackle the challenge of learning with limited data by exploring various solutions, including transfer learning from large datasets and improv- ing data efficiency on the existing training datasets. In Chapter 2, we propose a method based on transferable reward for query object localization to enable test-time policy adap- tation in new environments with limited labeled data. Chapter 3 proposes a solution to the problem of limited data for one modality in a multi-modal training set. Specifically, the chapter introduces a multi-modal model for new product recommendations that effectively leverages information from other modalities. By combining the static inherent properties of items with the limited customer engagement data (e.g., views, purchases, reviews, and likes), this approach enables efficient learning and recommendation of new products. In Chapter 4, we focus on real-world applications in fiber optic sensing, specifically vehicle run-off-road and manhole open/close events detection. Despite small training set sizes due to data collection constraints, we demonstrate how detecting these events can be achieved by investigating the underlying structure characterized by transformations and temporal relations, rather than simple pattern recognition.Ph.D.Includes bibliographical reference
The Connection between Buddhist Temples, the Landscape, and Monarchical Power: A Comparison between Tuoba Hong (471–499) from the Northern Wei Dynasty and Li Shimin (626–649) from the Tang Dynasty
No full textSince the Western Jin period, Buddhism has confronted bureaucratic power within the architectural landscape. In this study, historical records and archaeological reports of two Buddhist temples—the Siyuan Temple 思遠佛寺 built during the reign of Tuoba Hong 拓跋宏 and Emperor Xiaowen 孝文帝 of the Northern Wei dynasty, and the Zhaoren Temple 昭仁寺, built during the reign of Li Shimin 李世民 and Emperor Taizong 唐太宗 of the Tang dynasty—were examined. A comparison was made of the two temples’ geographic locations in relation to cities while considering period-specific phenomena. This study also considers mountains, water, and topographical features. The geographic information reflects differences in the ideas of the ruling class and monarchs of the two historical periods. The findings are that both Buddhist temples were close to the capital and both emperors demonstrated the supremacy of their power by building them. Therefore, the religious landscape owes its formation, development, and underlying significance to emperors and social groups
Author Correction: Mechanism of skull base osteoradionecrosis explored through laboratory assessment with propensity score-matched analysis
No full textTwo Transient Quasi-periodic Oscillations in -Ray Emission from the Blazar S4 0954+658
No full textIn this work, we report periodicity search analyses in the gamma-ray light
curve of the blazar S4 0954+658 monitoring undertaken by the Fermi Large Area
Telescope (LAT). Four analytical methods and a tool are adopted to detect any
periodic flux modulation and corresponding significance level, revealing that
(i) a 66 d quasi-periodic oscillation (QPO) with the significance level of spanning over 600 d from 2015 to 2016 (MJD 57145--57745), resulting in
continuous observation of nine cycles, which is one of the highest cycles
discerned in blazar gamma-ray light curve; (ii) a possible QPO of 210 d at a
moderate significance of lasted for over 880 d from 2020 to
2022 (MJD 59035--59915), which lasted for four cycles. In addition, we discuss
several physical models to explain the origin of the two transient QPOs and
conclude that a geometrical scenario involving a plasma blob moving helically
inside the jet can explain the time scale of the QPO.Comment: 10 pages, 11 figures; accepted for publication in Ap
Actuator failure assessment in smart composite laminates via principal component analysis
No full textIn this work, we propose a principal component analysis and system identification–based failure assessment approach for evaluating the partial actuator debonding failures in smart composite structures. Actuator debonding failure changes the structural dynamic characteristics and reduces the actuation capabilities as well in smart composite structures. First, the modeling of actuator debonding in smart composite laminate is developed using the finite element method, which incorporates the improved layerwise theory and higher-order electric potential field for the electromechanical coupling. Second, the structural responses obtained from the developed modeling are fed into the system identification to identify the system parameters of both healthy and damaged systems. Third, the achieved system parameters are further used for the statistical analysis by principal component analysis to extract the failure-sensitive features. Finally, a numerical example is studied using a 16-layer cross-ply laminate ([0/90]4s) as the substrate with various actuator debonding sizes. The results show that the actuator debonding failures can be well assessed, and the failure intensity and location can also be evaluated using the proposed approach
Multiwavelength Variability Analysis of the Blazar PKS 0727-11: An ~ 168 day Quasiperiodic Oscillation in the γ-Ray
No full textWe performed variability analysis of the multiwavelength light curves (LCs) for the flat-spectrum radio quasar PKS 0727-11. Using the generalized Lomb–Scargle periodogram, we identified a possible quasiperiodic oscillation (QPO) of ~168.6 days (persisted for six cycles, with a significance of 3.8 σ ) in the γ -ray LC during the flare period (MJD 54687–55738). It is the first time that periodic variations have been detected in this source, and further supported by other methods: weighted wavelet z -transform, phase dispersion minimization, REDFIT, autoregressive integrated moving average model, and structure function analysis. Cross-correlation analysis shows that there is a strong correlation between multiband light variations, indicating that γ -ray and radio flares may originate from the same disturbance, and the distance between the emission regions of γ -ray and radio flares is calculated based on the time lag. We demonstrate that QPO arising from the non-ballistic helical jet motion driven by the orbital motion in a supermassive binary black hole is a plausible physical explanation. In this scenario, the estimated mass of the primary black hole is M ∼ 3.66 × 10 ^8 –5.79 × 10 ^9 M _⊙
Acetylation of alcohols and phenols with acetic anhydride under solvent-free conditions using an ionic liquid based on morpholine as a recoverable and reusable catalyst
No full text
