Search

Sentiment analysis, an increasingly vital field in both academia and industry, plays a pivotal role in machine learning applications, particularly on social media platforms like Reddit. However, the efficacy of sentiment analysis models is hindered by the lack of expansive and fine-grained emotion datasets. To address this gap, our study leverages the GoEmotions dataset, comprising a diverse range of emotions, to evaluate sentiment analysis methods across a substantial corpus of 58,000 comments. Distinguished from prior studies by the Google team, which limited their analysis to only two models, our research expands the scope by evaluating a diverse array of models. We investigate the performance of traditional classifiers such as Naive Bayes and Support Vector Machines (SVM), as well as state-of-the-art transformer-based models including BERT, RoBERTa, and GPT. Furthermore, our evaluation criteria extend beyond accuracy to encompass nuanced assessments, including hierarchical classification based on varying levels of granularity in emotion categorization. Additionally, considerations such as computational efficiency are incorporated to provide a comprehensive evaluation framework. Our findings reveal that the RoBERTa model consistently outperforms the baseline models, demonstrating superior accuracy in fine-grained sentiment classification tasks. This underscores the substantial potential and significance of the RoBERTa model in advancing sentiment analysis capabilities., Comment: 11 pages, 5 figures, to be published in Computational and Experimental Simulations in Engineering - Proceedings of ICCES 2024 - Volume 2

Infrared excess is an important probe of sub-stellar companions and/or debris disks around white dwarfs (WDs). Such systems are still rare for in-depth understanding of their formation and long-term evolution. One of the largest spectroscopic surveys carried out by the Large sky Area Multi-Object fiber Spectroscopic Telescope (LAMOST) recently released more than $3000$ WDs, a significant fraction of which have not undergone excess search. Here we present cross-correlation of LAMOST DR5 WD catalog with the Pan-STARRS, SDSS, UKIDSS, 2MASS, and {\it WISE}. By performing SED (spectral energy distribution) fitting for 846 WDs with $WISE$ detections, we identify 50 candidates with infrared excess, including 7 candidate WD+M dwarf binaries, 31 candidate WD+brown dwarf (BD) binaries and 12 candidate WD+dust disk systems. 8 of the dust disk systems are our new identifications. Utilizing a systematic survey with accurate stellar parameters derived from spectral fitting, our work is an important addition to previous searches for infrared excess from SDSS and {\it Gaia} WDs, and provides a significant ($\gtrsim8\%$) complement to current database of WDs with candidate BD companions and dust disks. The frequencies of WD+BD binaries and WD+dust disk systems are constrained to be $\lesssim3.7\%$ and $\sim1.4\%$, respectively. The properties of candidate dust disk systems are discussed. All of our candidates require follow-up observations for confirmation owing to limited spatial resolution of {\it WISE}., Comment: 15 pages, 7 figures, 3 tables; accepted for publication in ApJ

Currently end-to-end deep learning based open-domain dialogue systems remain black box models, making it easy to generate irrelevant contents with data-driven models. Specifically, latent variables are highly entangled with different semantics in the latent space due to the lack of priori knowledge to guide the training. To address this problem, this paper proposes to harness the generative model with a priori knowledge through a cognitive approach involving mesoscopic scale feature disentanglement. Particularly, the model integrates the macro-level guided-category knowledge and micro-level open-domain dialogue data for the training, leveraging the priori knowledge into the latent space, which enables the model to disentangle the latent variables within the mesoscopic scale. Besides, we propose a new metric for open-domain dialogues, which can objectively evaluate the interpretability of the latent space distribution. Finally, we validate our model on different datasets and experimentally demonstrate that our model is able to generate higher quality and more interpretable dialogues than other models.

Purpose: To explore efficient encoding schemes for quantitative magnetization transfer (qMT) imaging with few constraints on model Theory and Methods: We combine two recently proposed models in a Bloch-McConnell equation: the dynamics of the free spin pool are confined to the hybrid state, and the dynamics of the semi-solid spin pool are described by the generalized Bloch model. We numerically optimize the flip angles and durations of a train of radio frequency pulses to enhance the encoding of three qMT parameters while accounting for all 8 parameters of the 2-pool model. We sparsely sample each time frame along this spin dynamics with a 3D radial koosh-ball trajectory, reconstruct the data with subspace modeling, and fit the qMT model with a neural network for computational efficiency. Results: We extracted qMT parameter maps of the whole brain with an effective resolution of 1.24mm from a 12.6-minute scan. In lesions of multiple sclerosis subjects, we observe a decreased size of the semi-solid spin pool and longer relaxation times, consistent with previous Conclusion: The encoding power of the hybrid state, combined with regularized image reconstruction, and the accuracy of the generalized Bloch model provide an excellent basis for efficient quantitative magnetization transfer imaging with few constraints on model parameters.

Neural networks are increasingly used to estimate parameters in quantitative MRI, in particular in magnetic resonance fingerprinting. Their advantages over the gold standard non-linear least square fitting are their superior speed and their immunity to the non-convexity of many fitting problems. We find, however, that in heterogeneous parameter spaces, i.e. in spaces in which the variance of the estimated parameters varies considerably, good performance is hard to achieve and requires arduous tweaking of the loss function, hyper parameters, and the distribution of the training data in parameter space. Here, we address these issues with a theoretically well-founded loss function: the Cram\'er-Rao bound (CRB) provides a theoretical lower bound for the variance of an unbiased estimator and we propose to normalize the squared error with respective CRB. With this normalization, we balance the contributions of hard-to-estimate and not-so-hard-to-estimate parameters and areas in parameter space, and avoid a dominance of the former in the overall training loss. Further, the CRB-based loss function equals one for a maximally-efficient unbiased estimator, which we consider the ideal estimator. Hence, the proposed CRB-based loss function provides an absolute evaluation metric. We compare a network trained with the CRB-based loss with a network trained with the commonly used means squared error loss and demonstrate the advantages of the former in numerical, phantom, and in vivo experiments., Comment: Xiaoxia Zhang, Quentin Duchemin, and Kangning Liu contributed equally to this work

The existence of $\gtrsim10^9M_\odot$ supermassive black holes (SMBHs) at redshift $z>6$ raises the problem of how such SMBHs can grow up within the cosmic time ($<1$\,Gyr) from small seed BHs. In this letter, we use the observations of $14$ Quasars at $z>6.5$ with mass estimates to constrain their seeds and early growth, by self-consistently considering the spin evolution and the possibility of super-Eddington accretion. We find that spin plays an important role in the growth of early SMBHs, and the constraints on seed mass and super-Eddington accretion fraction strongly depend on the assumed accretion history. If the accretion is coherent with single (or a small number of) episode(s), leading to high spins for the majority of accretion time, then the SMBH growth is relatively slow; and if the accretion is chaotic with many episodes and in each episode the total accreted mass is much less than the SMBH mass, leading to moderate/low spins, then the growth is relatively fast. The constraints on the seed mass and super-Eddington accretion fraction are degenerate. A significant fraction ($\gtrsim0.1\%-1\%$ in linear scale but $\sim 3-4$ dex in logarithmic scale for $10^3-10^4 M_\odot$ seeds) of super-Eddington accretion is required if the seed mass is not $\gg10^{5}M_\odot$, and the requirements of high seed-mass and/or super-Eddington accretion fraction are moderately relaxed if the accretion is chaotic., Comment: 7 pages, 2 figures, 1 table. Accepted for publication in ApJL

The spin distribution of massive black holes (MBHs) contains rich information on their assembly history. However, only limited information can be extracted from currently available spin measurements of MBHs owing to the small sample size and large measurement uncertainties. Upcoming X-ray telescopes with improved spectral resolution and larger effective area are expected to provide new insights into the growth history of MBHs. Here we investigate, at a proof of concept level, how stringent constraints can be placed on the accretion history of MBHs by the spin measurements from future X-ray missions. We assume a toy model consisting of a two-phase accretion history composed of an initial coherent phase with a constant disk orientation, followed by a chaotic phase with random disk orientations in each accretion episode. By utilizing mock spin data generated from such models and performing Bayesian Markov Chain Monte Carlo simulations, we find that most accretion models of MBHs can be reconstructed provided that $\gtrsim100$ MBH spins are measured with an accuracy of $\lesssim0.1$. We also quantify the precision of the reconstructed parameters by adopting various combinations of sample sizes and spin accuracies, and find that the sample size is more crucial to model reconstruction once the spin accuracy reaches $\sim 0.1$. To some extent, a better spin accuracy will compensate for a small sample size and vice versa. Future X-ray missions such as the Advanced Telescope for High Energy Astrophysics and the enhanced X-ray Timing and Polarimetry mission, may provide spin measurements of $\gtrsim100$ MBHs with an uncertainty of $\sim0.04-0.1$ and will thus put strong constraints on the MBH growth history., Comment: 11 pages, 9 figures, 1 table

Accretion of tidally disrupted stars (TDSs) is expected to contribute significantly to the growth of massive black holes (MBHs) with mass $\sim 10^6 M_{\odot}$ in galactic centers. In this paper, we quantitatively investigate the effect of the TDS accretion on the spin evolution of these relatively small MBHs, by also considering the accretion of gas-clouds with (many) chaotic episodes. We find that the accretion of TDSs can play an important role or even a dominant role in shaping the spin distribution of $\sim 10^6 M_{\odot}$ MBHs, depending on the contribution fraction ($f_{\rm TDE}$) of the TDS accretion to the MBH growth. If $f_{\rm TDE}$ is as large as $\gtrsim 0.9$, most $\sim 10^6M_{\odot}$ MBHs have low spins ($|a| \lesssim 0.3$); if $f_{\rm TDE}$ is as small as $\lesssim 0.1$, most $\sim 10^6 M_{\odot}$ MBHs have high spins ($|a| \gtrsim 0.7$). We also find that (1) the fraction of highly spinning $\sim 10^6 M_{\odot}$ MBHs in the TDS accretion states is smaller than that in the gas-cloud (AGN) accretion states, which is a consequence of more rapid spin decrease during the period of consecutive TDS accretion than the spin increase during the AGN periods when the spin is large; (2) the fraction of retrograde spin accretion in the TDS accretion states is almost the same as that of prograde spin accretion, while it is negligible in the gas-cloud (AGN) accretion states. Current scarce sample of AGNs ($\sim 10^6 M_{\odot}$) with spin measurements hints an insignificant contribution from TDS accretion to MBH growth. Future measurements on spins of $\sim 10^6 M_{\odot}$ MBHs may provide stronger constraints on the importance of both AGN and TDS accretion states in their growth history., Comment: 15pages, 8 figures, to appear in the Astrophysical Journal

Radiative efficiency is an important physical parameter that describes the fraction of accretion material converted to radiative energy for accretion onto massive black holes (MBHs). With the simplest So{\l}tan argument, the radiative efficiency of MBHs can be estimated by matching the mass density of MBHs in the local universe to the accreted mass density by MBHs during AGN/QSO phases. In this paper, we estimate the local MBH mass density through a combination of various determinations of the correlations between the masses of MBHs and the properties of MBH host galaxies, with the distribution functions of those galaxy properties. We also estimate the total energy density radiated by AGNs and QSOs by using various AGN/QSO X-ray luminosity functions in the literature. We then obtain several hundred estimates of the mean radiative efficiency of AGNs/QSOs. Under the assumption that those estimates are independent of each other and free of systematic effects, we apply the median statistics as described by Gott et al.\cite{got01} and find the mean radiative efficiency of AGNs/QSOs is $\epsilon=0.105^{+0.006}_{-0.008}$, which is consistent with the canonical value $\sim0.1$. Considering that about $20\%$ Compton-thick objects may be missed from current available X-ray surveys, the true mean radiative efficiency may be actually $\sim 0.12$., Comment: 8 pages, 2 figures

The spin distribution of massive black holes (MBHs) contains rich information on the MBH growth history. In this paper, we investigate the spin evolution of MBHs by assuming that each MBH experiences two-phase accretion, with an initial phase of coherent-accretion via either the standard thin disc or super-Eddington disc, followed by a chaotic-accretion phase composed of many episodes with different disc orientations. If the chaotic-phase is significant to the growth of an MBH, the MBH spin quickly reaches the maximum value because of the initial coherent-accretion, then changes to a quasi-equilibrium state and fluctuates around a value mainly determined by the mean ratio of the disc to the MBH mass ($M_{\bullet}$) in the chaotic-accretion episodes, and further declines due to late chaotic-accretion if $M_\bullet \gtrsim (1-3) \times 10^8 M_\odot$. The turning point to this decline is determined by the equality of the disc warp radius and disc size. By matching the currently available spin measurements with mock samples generated from the two-phase model(s) on the spin-mass plane, we find that MBHs must experience significant chaotic-accretion phase with many episodes and the mass accreted in each episode is roughly 1-2 percent of M_bh or less. MBHs with $M_{\bullet}\gtrsim 10^8 M_{\odot}$ appear to have intermediate-to-high spins ($\sim 0.5-1$), while lighter MBHs have higher spins ($\gtrsim 0.8$). The best matches also infer that (1) the radiative efficiencies ($\eta$) of those active MBHs appear to slightly decrease with $M_{\bullet}$; however, the correlation between $\eta$ and $M_{\bullet}$, if any, is weak; (2) the mean radiative efficiency of active MBHs is $<\eta> \sim 0.09-0.15$, consistent with the global constraints., Comment: 19 pages, 17 figures; to appear in the Astrophysical Journal