Search

The vertical structure of debris disks provides clues about their dynamical evolution and the collision rate of the unseen planetesimals. Thanks to the ever-increasing angular resolution of contemporary instruments and facilities, we are beginning to constrain the scale height of a handful of debris disks, either at near-infrared or millimeter wavelengths. Nonetheless, this is often done for individual targets only. We present here the geometric modeling of eight disks close to edge-on, all observed with the same instrument (SPHERE) and using the same mode (dual-beam polarimetric imaging). Motivated by the presence of CO gas in two out of the eight disks, we then investigate the impact that gas can have on the scale height by performing N-body simulations including gas drag and collisions. We show that gas can quickly alter the dynamics of particles (both in the radial and vertical directions), otherwise governed by gravity and radiation pressure. We find that, in the presence of gas, particles smaller than a few tens of microns can efficiently settle toward the midplane at the same time as they migrate outward beyond the birth ring. For second generation gas ($M_\mathrm{gas} \leq 0.1$ $M_\oplus$), the vertical settling should be best observed in scattered light images compared to observations at millimeter wavelengths. But if the gas has a primordial origin ($M_\mathrm{gas} \geq 1$ $M_\oplus$), the disk will appear very flat both at near-infrared and sub-mm wavelengths. Finally, far beyond the birth ring, our results suggest that the surface brightness profile can be as shallow as $\sim -2.25$., Comment: Accepted in MNRAS, 18 pages, 11 Figures

To understand how the multitude of planetary systems that have been discovered come to be, we need to study systems at different evolutionary stages, with different central stars but also in different environments. The most challenging environment for planet formation may be the harsh UV radiation field of nearby massive stars which quickly erodes disks by external photo-evaporation. We have observed the AT Pyx system, located in the head of a cometary globule in the Gum Nebula, to search for signs of ongoing planet formation. We used the extreme adaptive optics imager VLT/SPHERE to observe AT Pyx in polarized light as well as total intensity in the J, H and K-band. Additionally we employed VLT/NACO to observe the system in the L-band. We resolve the disk around AT Pyx in scattered light across multiple wavelengths. We find an extended (>126 au) disk, with an intermediate inclination between 35 deg and 42 deg. The disk shows complex sub-structure and we identify 2 and possibly 3 spiral-like features. Depending on the precise geometry of the disk (which we can not unambiguously infer from our data) the disk may be eccentric with an eccentricity of ~0.16 or partially self-shadowed. The spiral features and possible eccentricity are both consistent with signatures of an embedded gas giant planet equal in mass to Jupiter. Our own observations can rule out brown dwarf companions embedded in the resolved disk, but are not sensitive enough to detect gas giants. AT Pyx is the first disk in a cometray globule in the Gum Nebula which is spatially resolved. By comparison with disks in the Orion Nebula Cluster we note that the extension of the disk may be exceptional for this environment if the external UV radiation field is comparable to other cometary globules in the region. The signposts of ongoing planet formation are intriguing and need to be followed up with higher sensitivity., Comment: 11 pages, 9 figures, accepted for publication in A&A

To date, nearly two hundred planet-forming disks have been imaged with high resolution. Our propensity to study bright and extended objects is however biasing our view of the disk demography. In this work, we contribute to alleviate this bias by analyzing fifteen disks targeted with VLT/SPHERE that look faint in scattered light. Sources were selected based on a low far-IR excess from the spectral energy distribution. The comparison with the ALMA images available for a few sources shows that the scattered light surveyed by these datasets is only detected from a small portion of the disk extent. The mild anti-correlation between the disk brightness and the near-IR excess demonstrates that these disks are self-shadowed: the inner disk rim intercepts much starlight and leaves the outer disk in penumbra. Based on the uniform distribution of the disk brightness in scattered light across all spectral types, self-shadowing would act similarly for inner rims at a different distance from the star. We discuss how the illumination pattern of the outer disk may evolve with time. Some objects in the sample are proposed to be at an intermediate stage toward bright disks from the literature with either no shadow or with sign of azimuthally confined shadows., Comment: Accepted for publication in A\&A

Explainable AI (XAI) is a promising means of supporting human-AI collaborations for high-stakes visual detection tasks, such as damage detection tasks from satellite imageries, as fully-automated approaches are unlikely to be perfectly safe and reliable. However, most existing XAI techniques are not informed by the understandings of task-specific needs of humans for explanations. Thus, we took a first step toward understanding what forms of XAI humans require in damage detection tasks. We conducted an online crowdsourced study to understand how people explain their own assessments, when evaluating the severity of building damage based on satellite imagery. Through the study with 60 crowdworkers, we surfaced six major strategies that humans utilize to explain their visual damage assessments. We present implications of our findings for the design of XAI methods for such visual detection contexts, and discuss opportunities for future research., Comment: Accepted at NeurIPS 2021 Workshop on Artificial Intelligence for Humanitarian Assistance and Disaster Response (AI+HADR 2021)

GQ Lup B is a young and accreting, substellar companion that appears to drive a spiral arm in the circumstellar disk of its host star. We report high-contrast imaging observations of GQ Lup B with VLT/NACO at 4-5 $\mu$m and medium-resolution integral field spectroscopy with VLT/MUSE. The optical spectrum is consistent with an M9 spectral type, shows characteristics of a low-gravity atmosphere, and exhibits strong H$\alpha$ emission. The $H-M'$ color is $\gtrsim$1 mag redder than field dwarfs with similar spectral types and a detailed analysis of the spectral energy distribution (SED) from optical to mid-infrared wavelengths reveals excess emission in the $L'$, NB4.05, and $M'$ bands. The excess flux is well described by a blackbody component with $T_\mathrm{disk} \approx 460$ K and $R_\mathrm{disk} \approx 65\,R_\mathrm{J}$ and is expected to trace continuum emission from small grains in a protolunar disk. We derive an extinction of $A_V \approx 2.3$ mag from the broadband SED with a suspected origin in the vicinity of the companion. We also combine 15 yr of astrometric measurements and constrain the mutual inclination with the circumstellar disk to $84 \pm 9$ deg, indicating a tumultuous dynamical evolution or a stellar-like formation pathway. From the measured H$\alpha$ flux and the estimated companion mass, $M_\mathrm{p} \approx 30\,M_\mathrm{J}$, we derive an accretion rate of $\dot{M} \approx 10^{-6.5}\,M_\mathrm{J}\,\mathrm{yr}^{-1}$. We speculate that the disk is in a transitional stage in which the assembly of satellites from a pebble reservoir has opened a central cavity while GQ Lup B is in the final stages of its formation., Comment: 28 pages, 11 figures, accepted for publication in AJ

SCExAO at the Subaru telescope is a visible and near-infrared high-contrast imaging instrument employing extreme adaptive optics and coronagraphy. The instrument feeds the near-infrared light (JHK) to the integral-field spectrograph CHARIS. The spectropolarimetric capability of CHARIS is enabled by a Wollaston prism and is unique among high-contrast imagers. We present a detailed Mueller matrix model describing the instrumental polarization effects of the complete optical path, thus the telescope and instrument. From measurements with the internal light source, we find that the image derotator (K-mirror) produces strongly wavelength-dependent crosstalk, in the worst case converting ~95% of the incident linear polarization to circularly polarized light that cannot be measured. Observations of an unpolarized star show that the magnitude of the instrumental polarization of the telescope varies with wavelength between 0.5% and 1%, and that its angle is exactly equal to the altitude angle of the telescope. Using physical models of the fold mirror of the telescope, the half-wave plate, and the derotator, we simultaneously fit the instrumental polarization effects in the 22 wavelength bins. Over the full wavelength range, our model currently reaches a total polarimetric accuracy between 0.08% and 0.24% in the degree of linear polarization. We propose additional calibration measurements to improve the polarimetric accuracy to <0.1% and plan to integrate the complete Mueller matrix model into the existing CHARIS post-processing pipeline. Our calibrations of CHARIS' spectropolarimetric mode will enable unique quantitative polarimetric studies of circumstellar disks and planetary and brown dwarf companions., Comment: 27 pages, 16 figures, submitted to SPIE Optics + Photonics 2021

[Full abstract in the paper] In recent years, protoplanetary disks with spiral structures have been detected in scattered light, millimeter continuum, and CO gas emission. The mechanisms causing these structures are still under debate. A popular scenario to drive the spiral arms is the one of a planet perturbing the material in the disk. However, if the disk is massive, gravitational instability is usually the favored explanation. Multiwavelength studies could be helpful to distinguish between the two scenarios. So far, only a handful of disks with spiral arms have been observed in both scattered light and millimeter continuum. We aim to perform an in-depth characterization of the protoplanetary disk morphology around WaOph 6 analyzing data obtained at different wavelengths, as well as to investigate the origin of the spiral features in the disk. We present the first near-infrared polarimetric observations of WaOph 6 obtained with SPHERE at the VLT and compare them to archival millimeter continuum ALMA observations. We traced the spiral features in both data sets and estimated the respective pitch angles. We discuss the different scenarios that can give rise to the spiral arms in WaOph 6. We tested the planetary perturber hypothesis by performing hydrodynamical and radiative transfer simulations to compare them with scattered light and millimeter continuum observations., Comment: Accepted for publication in A&A. 17 pages, 9 figures in the main text and 3 figures in the Appendix

Background: The evolution of symptoms over time is at the heart of understanding and treating mental disorders. However, a principled, quantitative framework explaining symptom dynamics remains elusive. Here, we propose a Network Control Theory of Psychopathology allowing us to formally derive a theoretical control energy which we hypothesize quantifies resistance to future symptom improvement in Major Depressive Disorder (MDD). We test this hypothesis and investigate the relation to genetic and environmental risk as well as resilience. Methods: We modelled longitudinal symptom-network dynamics derived from N=2,059 Beck Depression Inventory measurements acquired over a median of 134 days in a sample of N=109 patients suffering from MDD. We quantified the theoretical energy required for each patient and time-point to reach a symptom-free state given individual symptom-network topology (E 0 ) and 1) tested if E 0 predicts future symptom improvement and 2) whether this relationship is moderated by Polygenic Risk Scores (PRS) of mental disorders, childhood maltreatment experience, and self-reported resilience. Outcomes: We show that E 0 indeed predicts symptom reduction at the next measurement and reveal that this coupling between E 0 and future symptom change increases with higher genetic risk and childhood maltreatment while it decreases with resilience. Interpretation: Our study provides a mechanistic framework capable of predicting future symptom improvement based on individual symptom-network topology and clarifies the role of genetic and environmental risk as well as resilience. Our control-theoretic framework makes testable, quantitative predictions for individual therapeutic response and provides a starting-point for the theory-driven design of personalized interventions. Funding: German Research Foundation and Interdisciplinary Centre for Clinical Research, M\"unster, Comment: 21 pages, 2 figures

Background: A therapeutic intervention in psychiatry can be viewed as an attempt to influence the brain's large-scale, dynamic network state transitions underlying cognition and behavior. Building on connectome-based graph analysis and control theory, Network Control Theory is emerging as a powerful tool to quantify network controllability - i.e., the influence of one brain region over others regarding dynamic network state transitions. If and how network controllability is related to mental health remains elusive. Methods: From Diffusion Tensor Imaging data, we inferred structural connectivity and inferred calculated network controllability parameters to investigate their association with genetic and familial risk in patients diagnosed with major depressive disorder (MDD, n=692) and healthy controls (n=820). Results: First, we establish that controllability measures differ between healthy controls and MDD patients while not varying with current symptom severity or remission status. Second, we show that controllability in MDD patients is associated with polygenic scores for MDD and psychiatric cross-disorder risk. Finally, we provide evidence that controllability varies with familial risk of MDD and bipolar disorder as well as with body mass index. Conclusions: We show that network controllability is related to genetic, individual, and familial risk in MDD patients. We discuss how these insights into individual variation of network controllability may inform mechanistic models of treatment response prediction and personalized intervention-design in mental health., Comment: 24 pages, 1 figure

Development of the intelligent autonomous robot technology presupposes its anticipated beneficial effect on the individuals and societies. In the case of such disruptive emergent technology, not only questions of how to build, but also why to build and with what consequences are important. The field of ethics of intelligent autonomous robotic cars is a good example of research with actionable practical value, where a variety of stakeholders, including the legal system and other societal and governmental actors, as well as companies and businesses, collaborate bringing about shared view of ethics and societal aspects of technology. It could be used as a starting platform for the approaches to the development of intelligent autonomous robots in general, considering human-machine interfaces in different phases of the life cycle of technology - the development, implementation, testing, use and disposal. Drawing from our work on ethics of autonomous intelligent robocars, and the existing literature on ethics of robotics, our contribution consists of a set of values and ethical principles with identified challenges and proposed approaches for meeting them. This may help stakeholders in the field of intelligent autonomous robotics to connect ethical principles with their applications. Our recommendations of ethical requirements for autonomous cars can be used for other types of intelligent autonomous robots, with the caveat for social robots that require more research regarding interactions with the users. We emphasize that existing ethical frameworks need to be applied in a context-sensitive way, by assessments in interdisciplinary, multi-competent teams through multi-criteria analysis. Furthermore, we argue for the need of a continuous development of ethical principles, guidelines, and regulations, informed by the progress of technologies and involving relevant stakeholders., Comment: 11 pages, 1 figure, 1 table

The deviation between chronological age and age predicted from neuroimaging data has been identified as a sensitive risk-marker of cross-disorder brain changes, growing into a cornerstone of biological age-research. However, Machine Learning models underlying the field do not consider uncertainty, thereby confounding results with training data density and variability. Also, existing models are commonly based on homogeneous training sets, often not independently validated, and cannot be shared due to data protection issues. Here, we introduce an uncertainty-aware, shareable, and transparent Monte-Carlo Dropout Composite-Quantile-Regression (MCCQR) Neural Network trained on N=10,691 datasets from the German National Cohort. The MCCQR model provides robust, distribution-free uncertainty quantification in high-dimensional neuroimaging data, achieving lower error rates compared to existing models across ten recruitment centers and in three independent validation samples (N=4,004). In two examples, we demonstrate that it prevents spurious associations and increases power to detect accelerated brain-aging. We make the pre-trained model publicly available.

Context. Debris disks have commonly been studied around intermediate-mass stars. Their intense radiation fields are believed to efficiently remove the small dust grains that are constantly replenished by collisions. For lower-mass stars, in particular M-stars, the dust removal mechanism needs to be further investigated given the much weaker radiation field produced by these objects. Aims. We present new polarimetric observations of the nearly edge-on disk around the pre-main sequence M-type star GSC 07396-00759, taken with VLT/SPHERE IRDIS, with the aim to better understand the morphology of the disk, its dust properties, and the star-disk interaction via the stellar mass-loss rate. Methods. We model our observations to characterize the location and properties of the dust grains using the Henyey-Greenstein approximation of the polarized phase function and evaluate the strength of the stellar winds. Results. We find that the observations are best described by an extended and highly inclined disk ($i\approx 84.3\,^{\circ}\pm0.3$) with a dust distribution centered at a radius $r_{0}\approx107\pm2$ au. The polarized phase function $S_{12}$ is best reproduced by an anisotropic scattering factor $g\approx0.6$ and small micron-sized dust grains with sizes $s>0.3\,\mathrm{\mu}$m. We furthermore discuss some of the caveats of the approach and a degeneracy between the grain size and the porosity. Conclusions. Even though the radius of the disk may be over-estimated, our results suggest that using a given scattering theory might not be sufficient to fully explain key aspects such as the shape of the phase function, or the dust grain size. With the caveats in mind, we find that the average mass-loss rate of GSC 07396-00759 can be up to 500 times stronger than that of the Sun, supporting the idea that stellar winds from low-mass stars can evacuate small dust grains from the disk., Comment: 22 pages, 17 figures, 4 tables

A growing body of literature has proposed formal approaches to audit algorithmic systems for biased and harmful behaviors. While formal auditing approaches have been greatly impactful, they often suffer major blindspots, with critical issues surfacing only in the context of everyday use once systems are deployed. Recent years have seen many cases in which everyday users of algorithmic systems detect and raise awareness about harmful behaviors that they encounter in the course of their everyday interactions with these systems. However, to date little academic attention has been granted to these bottom-up, user-driven auditing processes. In this paper, we propose and explore the concept of everyday algorithm auditing, a process in which users detect, understand, and interrogate problematic machine behaviors via their day-to-day interactions with algorithmic systems. We argue that everyday users are powerful in surfacing problematic machine behaviors that may elude detection via more centrally-organized forms of auditing, regardless of users' knowledge about the underlying algorithms. We analyze several real-world cases of everyday algorithm auditing, drawing lessons from these cases for the design of future platforms and tools that facilitate such auditing behaviors. Finally, we discuss work that lies ahead, toward bridging the gaps between formal auditing approaches and the organic auditing behaviors that emerge in everyday use of algorithmic systems., Comment: To appear in CSCW 2021. The co-first authors and co-senior authors each contributed equally to this work

The development of educational AI (AIEd) systems has often been motivated by their potential to promote educational equity and reduce achievement gaps across different groups of learners -- for example, by scaling up the benefits of one-on-one human tutoring to a broader audience, or by filling gaps in existing educational services. Given these noble intentions, why might AIEd systems have inequitable impacts in practice? In this chapter, we discuss four lenses that can be used to examine how and why AIEd systems risk amplifying existing inequities. Building from these lenses, we then outline possible paths towards more equitable futures for AIEd, while highlighting debates surrounding each proposal. In doing so, we hope to provoke new conversations around the design of equitable AIEd, and to push ongoing conversations in the field forward., Comment: The co-first authors contributed equally to this work

Recent work has explored how complementary strengths of humans and artificial intelligence (AI) systems might be productively combined. However, successful forms of human-AI partnership have rarely been demonstrated in real-world settings. We present the iterative design and evaluation of Lumilo, smart glasses that help teachers help their students in AI-supported classrooms by presenting real-time analytics about students' learning, metacognition, and behavior. Results from a field study conducted in K-12 classrooms indicate that students learn more when teachers and AI tutors work together during class. We discuss implications of this research for the design of human-AI partnerships. We argue for more participatory approaches to research and design in this area, in which practitioners and other stakeholders are deeply, meaningfully involved throughout the process. Furthermore, we advocate for theory-building and for principled approaches to the study of human-AI decision-making in real-world contexts., Comment: To appear in AI Magazine

Age prediction based on Magnetic Resonance Imaging (MRI) data of the brain is a biomarker to quantify the progress of brain diseases and aging. Current approaches rely on preparing the data with multiple preprocessing steps, such as registering voxels to a standardized brain atlas, which yields a significant computational overhead, hampers widespread usage and results in the predicted brain-age to be sensitive to preprocessing parameters. Here we describe a 3D Convolutional Neural Network (CNN) based on the ResNet architecture being trained on raw, non-registered T$_ 1$-weighted MRI data of N=10,691 samples from the German National Cohort and additionally applied and validated in N=2,173 samples from three independent studies using transfer learning. For comparison, state-of-the-art models using preprocessed neuroimaging data are trained and validated on the same samples. The 3D CNN using raw neuroimaging data predicts age with a mean average deviation of 2.84 years, outperforming the state-of-the-art brain-age models using preprocessed data. Since our approach is invariant to preprocessing software and parameter choices, it enables faster, more robust and more accurate brain-age modeling.

We present high-precision photometry and polarimetry for the protoplanetary disk around HD142527, with a focus on determining the light scattering parameters of the dust. We re-reduced polarimetric differential imaging data of HD142527 in the VBB (735 nm) and H-band (1625 nm) from the ZIMPOL and IRDIS subinstruments of SPHERE/VLT. With polarimetry and photometry based on reference star differential imaging, we were able to measure the linearly polarized intensity and the total intensity of the light scattered by the circumstellar disk with high precision. We used simple Monte Carlo simulations of multiple light scattering by the disk surface to derive constraints for three scattering parameters of the dust: the maximum polarization of $P_{\rm max}$, the asymmetry parameter $g$, and the single-scattering albedo $\omega$. We measure a reflected total intensity of $51.4\pm1.5$ mJy and $206\pm12$ mJy and a polarized intensity of $11.3\pm0.3$ mJy and $55.1\pm3.3$ mJy in the VBB and H-band, respectively. We also find in the visual range a degree of polarization that varies between $28\%$ on the far side of the disk and $17\%$ on the near side. The disk shows a red color for the scattered light intensity and the polarized intensity, which are about twice as high in the near-infrared when compared to the visual. We determine with model calculations the scattering properties of the dust particles and find evidence for strong forward scattering ($g\approx 0.5-0.75$), relatively low single-scattering albedo ($\omega \approx 0.2-0.5$), and high maximum polarization ($P_{\rm max} \approx 0.5-0.75$) at the surface on the far side of the disk for both observed wavelengths. The optical parameters indicate the presence of large aggregate dust particles, which are necessary to explain the high maximum polarization, the strong forward-scattering nature of the dust, and the observed red disk color., Comment: 20 pages, 14 figures