Search

In this paper, we present a semi-empirical calibration between the oxygen abundance and the $N2$ emission-line ratio for Low Ionization Nuclear Emission Regions (LINERs). This relation was derived by comparing the optical spectroscopic data of 118 nuclear spaxels classified as LINERs using three different BPT diagrams from the Mapping Nearby Galaxies survey (MaNGA) and sub-classified as weak (wAGN, 84 objects) and strong (sAGN, 34 objects) AGN (active galactic nucleus) from the WHAN diagnostic diagram and photoionization model results obtained with the {\sc cloudy} code assuming gas accretion into a black hole (representing an AGN). We found that our wAGN LINERs exhibit an oxygen abundance in the range of $8.50 \lesssim \mathrm{12+\log(O/H)} \lesssim 8.90 $, with an average value of $\mathrm{12+\log(O/H)}=8.68$, while our sAGN LINERs exhibit an oxygen abundance in the range of $8.51 \lesssim \: \mathrm{12+\log(O/H)} \: \lesssim \: 8.81 $, with an average value of $\mathrm{12+\log(O/H)}=8.65$. Our abundance estimations are in good agreement with those derived for another two different samples one of them with 463 Seyfert 2 objects and the other with 43 LINERs galaxies ionized by post-AGB stars, showing that the assumptions of our models are likely suitable for wAGN and sAGN LINERs. A relation between the equivalent width of the observed H$\alpha$ emission-line and the estimated ionization parameter provided by models was obtained. Our results also suggest that LINERs does not show a clear correlation between oxygen abundances and the stellar mass of the hosting galaxies., Comment: Accepted for publication in PASA

Progressive dimensionality reduction algorithms allow for visually investigating intermediate results, especially for large data sets. While different algorithms exist that progressively increase the number of data points, we propose an algorithm that allows for increasing the number of dimensions. Especially in spatio-temporal data, where each spatial location can be seen as one data point and each time step as one dimension, the data is often stored in a format that supports quick access to the individual dimensions of all points. Therefore, we propose Progressive Glimmer, a progressive multidimensional scaling (MDS) algorithm. We adapt the Glimmer algorithm to support progressive updates for changes in the data's dimensionality. We evaluate Progressive Glimmer's embedding quality and runtime. We observe that the algorithm provides more stable results, leading to visually consistent results for progressive rendering and making the approach applicable to streaming data. We show the applicability of our approach to spatio-temporal simulation ensemble data where we add the individual ensemble members progressively., Comment: 6 pages, 5 figures, presented at 2024 IEEE VIS Workshop on Progressive Data Analysis and Visualization (PDAV)

The datasets of most image quality assessment studies contain ratings on a categorical scale with five levels, from bad (1) to excellent (5). For each stimulus, the number of ratings from 1 to 5 is summarized and given in the form of the mean opinion score. In this study, we investigate families of multinomial probability distributions parameterized by mean and variance that are used to fit the empirical rating distributions. To this end, we consider quantized metric models based on continuous distributions that model perceived stimulus quality on a latent scale. The probabilities for the rating categories are determined by quantizing the corresponding random variables using threshold values. Furthermore, we introduce a novel discrete maximum entropy distribution for a given mean and variance. We compare the performance of these models and the state of the art given by the generalized score distribution for two large data sets, KonIQ-10k and VQEG HDTV. Given an input distribution of ratings, our fitted two-parameter models predict unseen ratings better than the empirical distribution. In contrast to empirical ACR distributions and their discrete models, our continuous models can provide fine-grained estimates of quantiles of quality of experience that are relevant to service providers to satisfy a target fraction of the user population., Comment: 5 pages

Current research provides methods to communicate uncertainty and adapts classical algorithms of the visualization pipeline to take the uncertainty into account. Various existing visualization frameworks include methods to present uncertain data but do not offer transformation techniques tailored to uncertain data. Therefore, we propose a software package for uncertainty-aware data analysis in Python (UADAPy) offering methods for uncertain data along the visualization pipeline. We aim to provide a platform that is the foundation for further integration of uncertainty algorithms and visualizations. It provides common utility functionality to support research in uncertainty-aware visualization algorithms and makes state-of-the-art research results accessible to the end user. The project is available at https://github.com/UniStuttgart-VISUS/uadapy.

For the Bio+Med-Vis Challenge 2024, we propose a visual analytics system as a redesign for the scatter pie chart visualization of cell type proportions of spatial transcriptomics data. Our design uses three linked views: a view of the histological image of the tissue, a stacked bar chart showing cell type proportions of the spots, and a scatter plot showing a dimensionality reduction of the multivariate proportions. Furthermore, we apply a compositional data analysis framework, the Aitchison geometry, to the proportions for dimensionality reduction and $k$-means clustering. Leveraging brushing and linking, the system allows one to explore and uncover patterns in the cell type mixtures and relate them to their spatial locations on the cellular tissue. This redesign shifts the pattern recognition workload from the human visual system to computational methods commonly used in visual analytics. We provide the code and setup instructions of our visual analytics system on GitHub (https://github.com/UniStuttgart-VISUS/va-for-spatial-transcriptomics)., Comment: 2 pages, 2 figures, challenge submission accepted by Bio+Med+Vis workshop at IEEE VIS 2024

Dimensionality reduction (DR) is a well-established approach for the visualization of high-dimensional data sets. While DR methods are often applied to typical DR benchmark data sets in the literature, they might suffer from high runtime complexity and memory requirements, making them unsuitable for large data visualization especially in environments outside of high-performance computing. To perform DR on large data sets, we propose the use of out-of-sample extensions. Such extensions allow inserting new data into existing projections, which we leverage to iteratively project data into a reference projection that consists only of a small manageable subset. This process makes it possible to perform DR out-of-core on large data, which would otherwise not be possible due to memory and runtime limitations. For metric multidimensional scaling (MDS), we contribute an implementation with out-of-sample projection capability since typical software libraries do not support it. We provide an evaluation of the projection quality of five common DR algorithms (MDS, PCA, t-SNE, UMAP, and autoencoders) using quality metrics from the literature and analyze the trade-off between the size of the reference set and projection quality. The runtime behavior of the algorithms is also quantified with respect to reference set size, out-of-sample batch size, and dimensionality of the data sets. Furthermore, we compare the out-of-sample approach to other recently introduced DR methods, such as PaCMAP and TriMAP, which claim to handle larger data sets than traditional approaches. To showcase the usefulness of DR on this large scale, we contribute a use case where we analyze ensembles of streamlines amounting to one billion projected instances.

Transmission spectroscopy has been a workhorse technique over the past two decades to constrain the physical and chemical properties of exoplanet atmospheres. One of its classical key assumptions is that the portion of the atmosphere it probes -- the terminator region -- is homogeneous. Several works in the past decade, however, have put this into question for highly irradiated, hot ($T_{eq}\gtrsim 1000$ K) gas giant exoplanets both empirically and via 3-dimensional modelling. While models predict clear differences between the evening (day-to-night) and morning (night-to-day) terminators, direct morning/evening transmission spectra in a wide wavelength range has not been reported for an exoplanet to date. Under the assumption of precise and accurate orbital parameters on WASP-39 b, here we report the detection of inhomogeneous terminators on the exoplanet WASP-39 b, which allows us to retrieve its morning and evening transmission spectra in the near-infrared ($2-5\ \mu$m) using JWST. We observe larger transit depths in the evening which are, on average, $405 \pm 88$ ppm larger than the morning ones, also having qualitatively larger features than the morning spectrum. The spectra are best explained by models in which the evening terminator is hotter than the morning terminator by $177^{+65}_{-57}$ K with both terminators having C/O ratios consistent with solar. General circulation models (GCMs) predict temperature differences broadly consistent with the above value and point towards a cloudy morning terminator and a clearer evening terminator., Comment: Published in Nature at https://www.nature.com/articles/s41586-024-07768-4. All code to produce plots (with data) can be found at https://github.com/nespinoza/wasp39-terminators

For the first time, a calibration between the HeI $\lambda5876$/H$\beta$ emission line ratio and the helium abundance $y$=12+log(He/H) for Narrow line regions (NLRs) of Seyfert~2 Active Galactic Nuclei (AGN) is proposed. In this context, observational data (taken from the SDSS-DR15 and from the literature) and direct abundance estimates (via the $T_{\rm e}$-method) for a sample of 65 local ($z \: < \: 0.2$) Seyfert~2 nuclei are considered. The resulting calibration estimates the $y$ abundance with an average uncertainty of 0.02 dex. Applying our calibration to spectroscopic data containing only strong emission lines, it yields a helium abundance distribution similar to that obtained via the $T_{\rm e}$-method. Some cautions must be considered to apply our calibration for Seyfert~2 nuclei with high values of electron temperature ($\gtrsim\: 20\,000$ K) or ionization parameter ($\log U > -2.0$)., Comment: Accepted for publication as Letter in MNRAS

Scale has become a main ingredient in obtaining strong machine learning models. As a result, understanding a model's scaling properties is key to effectively designing both the right training setup as well as future generations of architectures. In this work, we argue that scale and training research has been needlessly complex due to reliance on the cosine schedule, which prevents training across different lengths for the same model size. We investigate the training behavior of a direct alternative -- constant learning rate and cooldowns -- and find that it scales predictably and reliably similar to cosine. Additionally, we show that stochastic weight averaging yields improved performance along the training trajectory, without additional training costs, across different scales. Importantly, with these findings we demonstrate that scaling experiments can be performed with significantly reduced compute and GPU hours by utilizing fewer but reusable training runs. Our code is available at \url{https://github.com/epfml/schedules-and-scaling/}., Comment: Spotlight at NeurIPS 2024

We derive the nitrogen and oxygen abundances in the Narrow Line Regions (NLRs) of a sample of 38 local ($z \: < \: 0.4$) Seyfert~2 nuclei. For that, we consider narrow optical emission line intensities and direct estimates of the electron temperatures ($T_{\rm e}$-method). We find nitrogen abundances in the range $7.6 \: < \: \rm 12+log(N/H) \: < \: 8.6$ (mean value $8.06\pm0.22$) or $\rm 0.4 \: < \: (N/N_{\odot}) \: < 4.7$, in the metallicity regime $8.3 \: < \: \rm 12+log(O/H) \: < \: 9.0$. Our results indicate that the dispersion in N/H abundance for a fixed O/H value in AGNs is in agreement with that for disc \ion{H}{ii} regions with similar metallicity. We show that Seyfert~2 nuclei follow a similar (N/O)-(O/H) relation to the one followed by star-forming objects. Finally, we find that active galaxies called as 'nitrogen-loud' observed at very high redshift ($z \: > \: 5$) show N/O values in consonance with those derived for local NLRs. This result indicates that the main star-formation event is completed in the early evolution stages of active galaxies., Comment: Submitted to MNRAS. Comments are welcome!

In this work, we investigated the nitrogen and oxygen abundances in a sample of galaxies with Low Ionization Nuclear Emission Regions (LINERs) in their nucleus. Optical spectroscopic data (3 600 - 10 000 {\AA}) of 40 LINERs from the Mapping Nearby Galaxies (MaNGA) survey were considered. Only objects classified as retired galaxies, i.e. whose main ionization sources are post-Asymptotic Giant Branch (pAGB) stars, were selected. The abundance estimates were obtained through detailed photoionization models built with the cloudy code to reproduce a set of observational emission line intensities ratios of the sample. Our results show that LINERs have oxygen and nitrogen abundances in the ranges of $ 8.0 \leq 12+\log(O/H) \leq 9.0$ (mean value $8.74\pm 0.27$) and $7.6 \leq 12+\log(N/H) \leq 8.5$ (mean value $8.05\pm 0.25$), respectively. About 70% of the sample have oversolar O/H and N/H abundances. Our abundance estimates are in consonance with those for Seyfert 2 nuclei and H ii regions with the highest metallicity, indicating that these distinct object classes show similar enrichment of the interstellar medium (ISM). The LINERs in our sample are located in the higher N/O region of the N/O versus O/H diagram, showing an expected negative correlation between these two parameters. These results suggest that these LINERs mainly exhibit a secondary nitrogen production and could be acting some other mechanisms that deviate them from the usual theoretical secondary nitrogen production curve and the H ii regions observations. However, we did not find any evidence in our data able to support the literature suggested mechanisms. On the other hand, our results show that LINERs do not present any correlation between the N/O abundances and the stellar masses of the hosting galaxies., Comment: Accepted for publication in MNRAS

Addressing female underrepresentation in leadership positions has become a key policy objective. However, little is known about the extent to which leadership appeals differently to women. Collecting new data from a large firm, I document that women are substantially less likely to apply for early-career promotions. Realized application patterns and large-scale surveys reveal the role of an understudied feature of promotions -- having to assume responsibility over a team -- which is less appealing to women. This gender difference is not accounted for by standard explanations, such as success likelihood or confidence, but is rather a product of common design features of leadership positions.

The blinking statistics of quantum emitters and their corresponding Markov models play an important role in high resolution microscopy of biological samples as well as in nano-optoelectronics and many other fields of science and engineering. Current methods for analyzing the blinking statistics like the full counting statistics or the Viterbi algorithm break down for low photon rates. We present an evaluation scheme that eliminates the need for both a minimum photon flux and the usual binning of photon events which limits the measurement bandwidth. Our approach is based on higher order spectra of the measurement record which we model within the recently introduced method of quantum polyspectra from the theory of continuous quantum measurements. By virtue of this approach we can determine on- and off-switching rates of a semiconductor quantum dot at light levels 1000 times lower than in a standard experiment and 20 times lower than achieved with a scheme from full counting statistics. Thus a very powerful high-bandwidth approach to the parameter learning task of single photon hidden Markov models has been established with applications in many fields of science.

We present DiffInfinite, a hierarchical diffusion model that generates arbitrarily large histological images while preserving long-range correlation structural information. Our approach first generates synthetic segmentation masks, subsequently used as conditions for the high-fidelity generative diffusion process. The proposed sampling method can be scaled up to any desired image size while only requiring small patches for fast training. Moreover, it can be parallelized more efficiently than previous large-content generation methods while avoiding tiling artifacts. The training leverages classifier-free guidance to augment a small, sparsely annotated dataset with unlabelled data. Our method alleviates unique challenges in histopathological imaging practice: large-scale information, costly manual annotation, and protective data handling. The biological plausibility of DiffInfinite data is evaluated in a survey by ten experienced pathologists as well as a downstream classification and segmentation task. Samples from the model score strongly on anti-copying metrics which is relevant for the protection of patient data.

For the first time, the sulphur abundance relative to hydrogen (S/H) in the Narrow Line Regions of a sample of Seyfert 2 nuclei (Sy 2s) has been derived via direct estimation of the electron temperature. Narrow emission line intensities from the SDSS DR17 [in the wavelength range 3000 < $\lambda$ < 9100] and from the literature for a sample of 45 nearby ($z$ < 0.08) Sy 2s were considered. Our direct estimates indicate that Sy 2s have similar temperatures in the gas region where most of the S+ ions are located in comparison with that of star-forming regions (SFs). However, Sy 2s present higher temperature values ($\sim$10000 K) in the region where most of the S++ ions are located relative to that of SFs. We derive the total sulphur abundance in the range of 6.2 < 12 + log(S/H) < 7.5, corresponding to 0.1-1.8 times the solar value. These sulphur abundance values are lower by $\sim$0.4 dex than those derived in SFs with similar metallicity, indicating a distinct chemical enrichment of the ISM for these object classes. The S/O values for our Sy 2 sample present an abrupt ($\sim$0.5 dex) decrease with increasing O/H for the high metallicity regime [12 + log(O/H) > 8.7)], what is not seen for the SFs. However, when our Sy 2 estimates are combined with those from a large sample of star-forming regions, we did not find any dependence between S/O and O/H., Comment: Accepted for publication in MNRAS. 19 pages, 11 figures, 7 tables

A contactless technique for direct time-resolved measurements of the full dynamics of the adiabatic temperature change in electrocaloric materials is introduced. The infrared radiation emitted by the electrocaloric sample is sensitively detected with $\mu$s time resolution and mK temperature resolution. We present time-resolved measurements of the electrocaloric effect up to kHz frequencies of the driving electric field and down to small field strengths. The simultaneous recording of transients for applied electric field and induced polarization gives a comprehensive view on the correlation of electrocaloric and ferroelectric properties. The technique can further be applied to the continuous measurement of fatigue for $> 10^6$ electric field cycles., Comment: 12 pages, 11 figures

In this paper, we present a deep learning segmentation approach to classify and quantify the two most prevalent primary liver cancers - hepatocellular carcinoma and intrahepatic cholangiocarcinoma - from hematoxylin and eosin (H&E) stained whole slide images. While semantic segmentation of medical images typically requires costly pixel-level annotations by domain experts, there often exists additional information which is routinely obtained in clinical diagnostics but rarely utilized for model training. We propose to leverage such weak information from patient diagnoses by deriving complementary labels that indicate to which class a sample cannot belong to. To integrate these labels, we formulate a complementary loss for segmentation. Motivated by the medical application, we demonstrate for general segmentation tasks that including additional patches with solely weak complementary labels during model training can significantly improve the predictive performance and robustness of a model. On the task of diagnostic differentiation between hepatocellular carcinoma and intrahepatic cholangiocarcinoma, we achieve a balanced accuracy of 0.91 (CI 95%: 0.86 - 0.95) at case level for 165 hold-out patients. Furthermore, we also show that leveraging complementary labels improves the robustness of segmentation and increases performance at case level.

Augmented reality (AR) games are a rich environment for researching and testing computational systems that provide subtle user guidance and training. In particular computer systems that aim to augment a user's situation awareness benefit from the range of sensors and computing power available in AR headsets. In this work-in-progress paper, we present a new environment for research into situation awareness and attention guidance (SAAG): an augmented reality version of the board game Carcassonne. We also present our initial work in producing a SAAG pipeline, including the creation of game state encodings, the development and training of a gameplay AI, and the design of situation modelling and gaze tracking systems.

The ionizing source of Low Ionization Nuclear Emission Regions (LINERs) is uncertain. Because of this, an empirical relation to determine the chemical abundances of these objects has not been proposed. In this work, for the first time, we derived two semi-empirical calibrations based on photoionization models to estimate the oxygen abundance of LINERS as a function of the $N2$ and $O3N2$ emission-line intensity ratios. These relations were calibrated using oxygen abundance estimations obtained by comparing the observational emission-line ratios of 43 LINER galaxies (taken from the MaNGA survey) and grids of photoionization models built with the {\sc Cloudy} code assuming post-Asymptotic Giant Branch (post-AGB) stars with different temperatures. We found that the oxygen abundance of LINERs in our sample is in the $\rm 8.48 <~ 12+log(O/H) <~ 8.84$ range, with a mean value of $\rm 12+\log(O/H)=8.65$. We recommend the use of the $N2$ index to estimate the oxygen abundances of LINERs, since the calibration with this index presented a much smaller dispersion than the $O3N2$ index. In addition, the estimated metallicities are in good agreement with those derived by extrapolating the disk oxygen abundance gradients to the centre of the galaxies showing that the assumptions of the models are suitable for LINERs. We also obtained a calibration between the logarithm of the ionization parameter and the [OIII]/[OII] emission-line ratio., Comment: 16 pages, 14 figures, 3 tables, accepted for publication in MNRAS

Most organizations rely on managers to identify talented workers. However, managers who are evaluated on team performance have an incentive to hoard workers. This study provides the first empirical evidence of talent hoarding using personnel records and survey evidence from a large manufacturing firm. Talent hoarding is reported by three-fourths of managers, is detectable in managerial decisions, and occurs more frequently when hoarding incentives are stronger. Using quasi-random variation in exposure to talent hoarding, I demonstrate that hoarding deters workers from applying to new positions, inhibiting worker career progression and altering the allocation of talent in the firm.

Inferring causal structures from experimentation is a central task in many domains. For example, in biology, recent advances allow us to obtain single-cell expression data under multiple interventions such as drugs or gene knockouts. However, the targets of the interventions are often uncertain or unknown and the number of observations limited. As a result, standard causal discovery methods can no longer be reliably used. To fill this gap, we propose a Bayesian framework (BaCaDI) for discovering and reasoning about the causal structure that underlies data generated under various unknown experimental or interventional conditions. BaCaDI is fully differentiable, which allows us to infer the complex joint posterior over the intervention targets and the causal structure via efficient gradient-based variational inference. In experiments on synthetic causal discovery tasks and simulated gene-expression data, BaCaDI outperforms related methods in identifying causal structures and intervention targets., Comment: Accepted to AISTATS 2023. 26 pages

In univariate data, there exist standard procedures for identifying dominating features that produce the largest observations. However, in the multivariate setting, the situation is quite different. This paper aims to provide tools and algorithms for detecting dominating directional components in multivariate data. We study general heavy-tailed multivariate random vectors in dimension $d\geq 2$ and present consistent estimators which can be used to evaluate why the data is heavy-tailed. This is done by identifying the set of the riskiest directional components. The results are of particular interest in insurance when setting reinsurance policies and in finance when hedging a portfolio of multiple assets., Comment: 24 pages, 5 figures

The analysis of a continuous measurement record $z(t)$ poses a fundamental challenge in quantum measurement theory. Different approaches have been used in the past as records can, e.g., exhibit predominantly Gaussian noise, telegraph noise, or clicks at random times. The last case may appear as photon clicks in an optical spin noise measurement at very low probe laser power. Here we show that such random-time quantum measurements can similarly to the first two cases be analyzed in terms of higher-order temporal correlations of the detector output $z(t)$ and be related to the Liouvillian of the measured quantum system. Our analysis in terms of up to fourth-order spectra (quantum polyspectra) shows that this new type of spectra reveals the same valuable information as previously studied higher-order spectra in case of usual continuous quantum measurements. Surprisingly, broad-band system dynamics is revealed even for deliberately low average measurement rates. Many applications are envisioned in high-resolution spectroscopy, single-photon microscopy, circuit quantum electrodynamics, quantum sensing, and quantum measurements in general.