Search

Understanding the chemistry of molecular clouds is pivotal to elucidate star formation and galaxy evolution. As one of the important molecular ions, HCNH+ plays an important role in this chemistry. Yet, its behavior and significance under extreme conditions, such as in the CMZs of external galaxies, are still largely unexplored. We aim to reveal the physical and chemical properties of the CMZ in the starburst galaxy NGC253 with multiple HCNH+ transitions to shed light on the molecule's behavior under the extreme physical conditions of a starburst. We employ molecular line data including results for four rotational transitions of HCNH+ from the ALCHEMI large program to investigate underlying physical and chemical processes. Despite weak intensities, HCNH+ emission is widespread throughout NGC253's CMZ, which suggests that this molecular ion can effectively trace large-scale structures within molecular clouds. Using the quantum mechanical coupled states approximation, we computed rate coefficients for collisions of HCNH+ with para-H2 and ortho-H2 at kinetic temperatures up to 500 K. Using these coefficients in a non-LTE modeling framework and employing a Monte Carlo Markov chain analysis, we find that HCNH+ emission originates from regions with H2 number densities of $\sim10^{2.80}-10^{3.55}$~cm$^{-3}$, establishing HCNH+ as a tracer of low-density environments. Our analysis reveals that most of the HCNH+ abundances in the CMZ of NGC253 are higher than all reported values in the Milky Way. We performed static, PDR, and shock modeling, and found that recurrent shocks could potentially account for the elevated HCNH+ abundances observed in this CMZ. We propose that the unexpectedly high HCNH+ abundances may result from chemical enhancement, primarily driven by the elevated gas temperatures and cosmic ray ionization rates of shocked, low-density gas in the nuclear starburst regions of NGC253., Comment: 18 pages, 14 figures, 2 tables, accepted for publication in A&A

Artificial intelligence (AI) has transformed digital pathology by enabling biomarker prediction from high-resolution whole slide images (WSIs). However, current methods are computationally inefficient, processing thousands of redundant tiles per WSI and requiring complex aggregator models. We introduce EAGLE (Efficient Approach for Guided Local Examination), a deep learning framework that emulates pathologists by selectively analyzing informative regions. EAGLE incorporates two foundation models: CHIEF for efficient tile selection and Virchow2 for extracting high-quality features. Benchmarking was conducted against leading slide- and tile-level foundation models across 31 tasks from four cancer types, spanning morphology, biomarker prediction and prognosis. EAGLE outperformed state-of-the-art foundation models by up to 23% and achieved the highest AUROC overall. It processed a slide in 2.27 seconds, reducing computational time by more than 99% compared to existing models. This efficiency enables real-time workflows, allows pathologists to validate all tiles which are used by the model during analysis, and eliminates dependence on high-performance computing, making AI-powered pathology more accessible. By reliably identifying meaningful regions and minimizing artifacts, EAGLE provides robust and interpretable outputs, supporting rapid slide searches, integration into multi-omics pipelines and emerging clinical foundation models.

The dynamical cavity method and its backtracking version provide a powerful approach to studying the properties of dynamical processes on large random graphs. This paper extends these methods to hypergraphs, enabling the analysis of interactions involving more than two variables. We apply them to analyse the $k$-XOR-satisfiability ($k$-XOR-SAT) problem, an important model in theoretical computer science which is closely related to the diluted $p$-spin model from statistical physics. In particular, we examine whether the quench dynamics -- a deterministic, locally greedy process -- can find solutions with only a few violated constraints on $d$-regular $k$-uniform hypergraphs. Our results demonstrate that the methods accurately characterize the attractors of the dynamics. It enables us to compute the energy reached by typical trajectories of the dynamical process in different parameter regimes. We show that these predictions are accurate, including cases where a classical mean-field approach fails., Comment: 26 pages

Investigating the assembly history of the most massive and passive galaxies will enhance our understanding of why galaxies exhibit such a remarkable diversity in structure and morphology. In this paper, we simultaneously investigate the assembly history and redshift evolution of semi-analytically modelled galaxy properties of central galaxies between 0.56 < z < 4.15, alongside their connection to their halos as a function of large-scale environment. We extract sub-samples of galaxies from a mock catalogue representative for the BOSS-CMASS sample, which includes the most massive and passively evolving system known today. Utilising typical galaxy properties such as star formation rate, (g-i) colour, or cold gas-phase metallicity (Zcold), we track the redshift evolution of these properties across the main progenitor trees. We present results on galaxy and halo properties, including their growth and clustering functions. Our findings indicate that galaxies in the highest stellar and halo mass regimes are least metal-enriched (using Zcold as a proxy) and consistently exhibit significantly larger black hole masses and higher clustering amplitudes compared to sub-samples selected by e.g. colour or star formation rate. This population forms later and also retains large reservoirs of cold gas. In contrast, galaxies in the intermediate and lower stellar/halo mass regimes consume their cold gas at higher redshift and were among the earliest and quickest to assemble. We observe a clear trend where the clustering of the galaxies selected according to their Zcold-values (either low-Zcold or high-Zcold) depends on the density of their location within the large-scale environment. We assume that in particular galaxies in the low/high-Zcold sub-samples form and evolve through distinct evolutionary channels, which are predetermined by their location within the large-scale environment of the cosmic web., Comment: 21 pages, 12 figures, accepted for publication in A&A

Recent approaches have successfully focused on the segmentation of static reconstructions, thereby equipping downstream applications with semantic 3D understanding. However, the world in which we live is dynamic, characterized by numerous interactions between the environment and humans or robotic agents. Static semantic maps are unable to capture this information, and the naive solution of rescanning the environment after every change is both costly and ineffective in tracking e.g. objects being stored away in drawers. With Lost & Found we present an approach that addresses this limitation. Based solely on egocentric recordings with corresponding hand position and camera pose estimates, we are able to track the 6DoF poses of the moving object within the detected interaction interval. These changes are applied online to a transformable scene graph that captures object-level relations. Compared to state-of-the-art object pose trackers, our approach is more reliable in handling the challenging egocentric viewpoint and the lack of depth information. It outperforms the second-best approach by 34% and 56% for translational and orientational error, respectively, and produces visibly smoother 6DoF object trajectories. In addition, we illustrate how the acquired interaction information in the dynamic scene graph can be employed in the context of robotic applications that would otherwise be unfeasible: We show how our method allows to command a mobile manipulator through teach & repeat, and how information about prior interaction allows a mobile manipulator to retrieve an object hidden in a drawer. Code, videos and corresponding data are accessible at https://behretj.github.io/LostAndFound., Comment: Webpage: https://behretj.github.io/LostAndFound

Incorporating AI technologies into digital infrastructure offers transformative potential for energy management, particularly in enhancing energy efficiency and supporting net-zero objectives. However, the complexity of IoT-generated datasets often poses a significant challenge, hindering the translation of research insights into practical, real-world applications. This paper presents the design of an interactive visualization tool, BiTSA. The tool enables building managers to interpret complex energy data quickly and take immediate, data-driven actions based on real-time insights. By integrating advanced forecasting models with an intuitive visual interface, our solution facilitates proactive decision-making, optimizes energy consumption, and promotes sustainable building management practices. BiTSA will empower building managers to optimize energy consumption, control demand-side energy usage, and achieve sustainability goals., Comment: 4 pages, 4 figures, 3 tables

We formulate a depth-averaged non-hydrostatic model to solve wave equations with generation by a moving bottom. This model is built upon the shallow water equations, which are widely used in tsunami wave modelling. An extension leads to two additional unknowns to be solved: vertical momentum and non-hydrostatic pressure. We show that a linear vertical velocity assumption turns out to give us a quadratic pressure relation, which is equivalent to Boussinesq-type equations. However, this extension involves a time derivative of an unknown parameter, rendering the solution by a projection method ambiguous. In this study, we derive an alternative form of the elliptic system of equations to avoid such ambiguity. The new set of equations satisfies the desired solubility property, while also consistently representing the non-flat moving topography wave generation. Validations are performed using several test cases based on previous experiments and a high-fidelity simulation. First, we show the efficiency of our model in solving a vertical movement, which represents an undersea earthquake-generated tsunami. Following that, we demonstrate the accuracy of the model for landslide-generated waves. Finally, we compare the performance of our novel set of equations with the linear and simplified quadratic pressure profiles.

The growing integration of digitized infrastructure with Internet of Things (IoT) networks has transformed the management and optimization of building energy consumption. By leveraging IoT-based monitoring systems, stakeholders such as building managers, energy suppliers, and policymakers can make data-driven decisions to improve energy efficiency. However, accurate energy forecasting and analytics face persistent challenges, primarily due to the inherent physical constraints of buildings and the diverse, heterogeneous nature of IoT-generated data. In this study, we conduct a comprehensive benchmarking of two publicly available IoT datasets, evaluating the performance of time series foundation models in the context of building energy analytics. Our analysis shows that single-modal models demonstrate significant promise in overcoming the complexities of data variability and physical limitations in buildings, with future work focusing on optimizing multi-modal models for sustainable energy management., Comment: 7 pages, 1 figures, and 4 tables

We survey the role of Lie algebras in the study of unstable homotopy groups., Comment: 17 pages v2: many minor corrections and clarifications following the suggestions of the referees

Why do biological and artificial neurons sometimes modularise, each encoding a single meaningful variable, and sometimes entangle their representation of many variables? In this work, we develop a theory of when biologically inspired networks -- those that are nonnegative and energy efficient -- modularise their representation of source variables (sources). We derive necessary and sufficient conditions on a sample of sources that determine whether the neurons in an optimal biologically-inspired linear autoencoder modularise. Our theory applies to any dataset, extending far beyond the case of statistical independence studied in previous work. Rather we show that sources modularise if their support is ``sufficiently spread''. From this theory, we extract and validate predictions in a variety of empirical studies on how data distribution affects modularisation in nonlinear feedforward and recurrent neural networks trained on supervised and unsupervised tasks. Furthermore, we apply these ideas to neuroscience data, showing that range independence can be used to understand the mixing or modularising of spatial and reward information in entorhinal recordings in seemingly conflicting experiments. Further, we use these results to suggest alternate origins of mixed-selectivity, beyond the predominant theory of flexible nonlinear classification. In sum, our theory prescribes precise conditions on when neural activities modularise, providing tools for inducing and elucidating modular representations in brains and machines., Comment: 40 pages, 16 figures. WD and KH contributed equally; LH and JHL contributed equally

Learning from demonstration is a promising way of teaching robots new skills. However, a central problem when executing acquired skills is to recognize risks and failures. This is essential since the demonstrations usually cover only a few mostly successful cases. Inevitable errors during execution require specific reactions that were not apparent in the demonstrations. In this paper, we focus on teaching the robot situational awareness from an initial skill demonstration via kinesthetic teaching and sparse labeling of autonomous skill executions as safe or risky. At runtime, our system, called ILeSiA, detects risks based on the perceived camera images by encoding the images into a low-dimensional latent space representation and training a classifier based on the encoding and the provided labels. In this way, ILeSiA boosts the confidence and safety with which robotic skills can be executed. Our experiments demonstrate that classifiers, trained with only a small amount of user-provided data, can successfully detect numerous risks. The system is flexible because the risk cases are defined by labeling data. This also means that labels can be added as soon as risks are identified by a human supervisor. We provide all code and data required to reproduce our experiments at imitrob.ciirc.cvut.cz/publications/ilesia., Comment: 7 pages, 8 figures

We use a neural network model and ALMA observations of HCN and HNC to constrain the physical conditions, most notably the cosmic-ray ionization rate (CRIR, zeta), in the Central Molecular Zone (CMZ) of the starburst galaxy NGC 253. Using output from the chemical code UCLCHEM, we train a neural network model to emulate UCLCHEM and derive HCN and HNC molecular abundances from a given set of physical conditions. We combine the neural network with radiative transfer modeling to generate modeled integrated intensities, which we compare to measurements of HCN and HNC from the ALMA Large Program ALCHEMI. Using a Bayesian nested sampling framework, we constrain the CRIR, molecular gas volume and column densities, kinetic temperature, and beam-filling factor across NGC 253's CMZ. The neural network model successfully recovers UCLCHEM molecular abundances with about 3 percent error and, when used with our Bayesian inference algorithm, increases the parameter inference speed tenfold. We create images of these physical parameters across NGC 253's CMZ at 50 pc resolution and find that the CRIR, in addition to the other gas parameters, is spatially variable with zeta a few times 10^{14} s^{-1} at greater than 100 pc from the nucleus, increasing to zeta greater than 10^{-13} s^{-1} at its center. These inferred CRIRs are consistent within 1 dex with theoretical predictions based on non-thermal emission. Additionally, the high CRIRs estimated in NGC 253's CMZ can be explained by the large number of cosmic-ray-producing sources as well as a potential suppression of cosmic-ray diffusion near their injection sites.

Connected Autonomous Vehicles have great potential to improve automobile safety and traffic flow, especially in cooperative applications where perception data is shared between vehicles. However, this cooperation must be secured from malicious intent and unintentional errors that could cause accidents. Previous works typically address singular security or reliability issues for cooperative driving in specific scenarios rather than the set of errors together. In this paper, we propose CONClave, a tightly coupled authentication, consensus, and trust scoring mechanism that provides comprehensive security and reliability for cooperative perception in autonomous vehicles. CONClave benefits from the pipelined nature of the steps such that faults can be detected significantly faster and with less compute. Overall, CONClave shows huge promise in preventing security flaws, detecting even relatively minor sensing faults, and increasing the robustness and accuracy of cooperative perception in CAVs while adding minimal overhead., Comment: 6 pages, 6 figures, Design Automation Conference June 2024

Advancements in artificial intelligence have driven the development of numerous pathology foundation models capable of extracting clinically relevant information. However, there is currently limited literature independently evaluating these foundation models on truly external cohorts and clinically-relevant tasks to uncover adjustments for future improvements. In this study, we benchmarked 19 histopathology foundation models on 13 patient cohorts with 6,818 patients and 9,528 slides from lung, colorectal, gastric, and breast cancers. The models were evaluated on weakly-supervised tasks related to biomarkers, morphological properties, and prognostic outcomes. We show that a vision-language foundation model, CONCH, yielded the highest performance when compared to vision-only foundation models, with Virchow2 as close second. The experiments reveal that foundation models trained on distinct cohorts learn complementary features to predict the same label, and can be fused to outperform the current state of the art. An ensemble combining CONCH and Virchow2 predictions outperformed individual models in 55% of tasks, leveraging their complementary strengths in classification scenarios. Moreover, our findings suggest that data diversity outweighs data volume for foundation models. Our work highlights actionable adjustments to improve pathology foundation models.

Stroke remains a leading cause of global morbidity and mortality, placing a heavy socioeconomic burden. Over the past decade, advances in endovascular reperfusion therapy and the use of CT and MRI imaging for treatment guidance have significantly improved patient outcomes and are now standard in clinical practice. To develop machine learning algorithms that can extract meaningful and reproducible models of brain function for both clinical and research purposes from stroke images - particularly for lesion identification, brain health quantification, and prognosis - large, diverse, and well-annotated public datasets are essential. While only a few datasets with (sub-)acute stroke data were previously available, several large, high-quality datasets have recently been made publicly accessible. However, these existing datasets include only MRI data. In contrast, our dataset is the first to offer comprehensive longitudinal stroke data, including acute CT imaging with angiography and perfusion, follow-up MRI at 2-9 days, as well as acute and longitudinal clinical data up to a three-month outcome. The dataset includes a training dataset of n = 150 and a test dataset of n = 100 scans. Training data is publicly available, while test data will be used exclusively for model validation. We are making this dataset available as part of the 2024 edition of the Ischemic Stroke Lesion Segmentation (ISLES) challenge (https://www.isles-challenge.org/), which continuously aims to establish benchmark methods for acute and sub-acute ischemic stroke lesion segmentation, aiding in creating open stroke imaging datasets and evaluating cutting-edge image processing algorithms.

The projected sensitivity of the effective electron neutrino-mass measurement with the KATRIN experiment is below 0.3 eV (90 % CL) after five years of data acquisition. The sensitivity is affected by the increased rate of the background electrons from KATRIN's main spectrometer. A special shifted-analysing-plane (SAP) configuration was developed to reduce this background by a factor of two. The complex layout of electromagnetic fields in the SAP configuration requires a robust method of estimating these fields. We present in this paper a dedicated calibration measurement of the fields using conversion electrons of gaseous $^\mathrm{83m}$Kr, which enables the neutrino-mass measurements in the SAP configuration., Comment: 19 pages, 11 figures

AI assistants are being increasingly used by students enrolled in higher education institutions. While these tools provide opportunities for improved teaching and education, they also pose significant challenges for assessment and learning outcomes. We conceptualize these challenges through the lens of vulnerability, the potential for university assessments and learning outcomes to be impacted by student use of generative AI. We investigate the potential scale of this vulnerability by measuring the degree to which AI assistants can complete assessment questions in standard university-level STEM courses. Specifically, we compile a novel dataset of textual assessment questions from 50 courses at EPFL and evaluate whether two AI assistants, GPT-3.5 and GPT-4 can adequately answer these questions. We use eight prompting strategies to produce responses and find that GPT-4 answers an average of 65.8% of questions correctly, and can even produce the correct answer across at least one prompting strategy for 85.1% of questions. When grouping courses in our dataset by degree program, these systems already pass non-project assessments of large numbers of core courses in various degree programs, posing risks to higher education accreditation that will be amplified as these models improve. Our results call for revising program-level assessment design in higher education in light of advances in generative AI., Comment: 20 pages, 8 figures

Many socio-economical critical domains (such as sustainability, public health, and disasters) are characterized by highly complex and dynamic systems, requiring data and model-driven simulations to support decision-making. Due to a large number of unknowns, decision-makers usually need to generate ensembles of stochastic scenarios, requiring hundreds or thousands of individual simulation instances, each with different parameter settings corresponding to distinct scenarios, As the number of model parameters increases, the number of potential timelines one can simulate increases exponentially. Consequently, simulation ensembles are inherently sparse, even when they are extremely large. This necessitates a platform for (a) deciding which simulation instances to execute and (b) given a large simulation ensemble, enabling decision-makers to explore the resulting alternative timelines, by extracting and visualizing consistent, yet diverse timelines from continuous-coupled simulation ensembles. In this article, we present DataStorm-EM platform for data- and model-driven simulation ensemble management, optimization, analysis, and exploration, describe underlying challenges and present our solution.

How do different architectural design choices influence the space of solutions that a transformer can implement and learn? How do different components interact with each other to shape the model's hypothesis space? We investigate these questions by characterizing the solutions simple transformer blocks can implement when challenged to solve the histogram task -- counting the occurrences of each item in an input sequence from a fixed vocabulary. Despite its apparent simplicity, this task exhibits a rich phenomenology: our analysis reveals a strong inter-dependence between the model's predictive performance and the vocabulary and embedding sizes, the token-mixing mechanism and the capacity of the feed-forward block. In this work, we characterize two different counting strategies that small transformers can implement theoretically: relation-based and inventory-based counting, the latter being less efficient in computation and memory. The emergence of either strategy is heavily influenced by subtle synergies among hyperparameters and components, and depends on seemingly minor architectural tweaks like the inclusion of softmax in the attention mechanism. By introspecting models trained on the histogram task, we verify the formation of both mechanisms in practice. Our findings highlight that even in simple settings, slight variations in model design can cause significant changes to the solutions a transformer learns., Comment: 33 pages, Accepted at Mechanistic Interpretability Workshop, ICML 2024

Building off of many recent advances in the subject by many different researchers, we describe a picture of A-equivariant chromatic homotopy theory which mirrors the now classical non-equivariant picture of Morava, Miller-Ravenel-Wilson, and Devinatz-Hopkins-Smith, where A is a finite abelian p-group. Specifically, we review the structure of the Balmer spectrum of the category of A-spectra, and the work of Hausmann-Meier connecting this to MU_A and equivariant formal group laws. Generalizing work of Bhattacharya-Guillou-Li, we introduce equivariant analogs of v_n-self maps, and generalizing work of Carrick and Balderrama, we introduce equivariant analogs of the chromatic tower, and give equivariant analogs of the smash product and chromatic convergence theorems. The equivariant monochromatic theory is also discussed. We explore computational examples of this theory in the case of A = C_2, where we connect equivariant chromatic theory with redshift phenomena in Mahowald invariants., Comment: 69 pages. Changes to v4: added reference to Hanke and Wiemeler's work on MU_C2 and corrected a typo

The fact that neutrinos carry a non-vanishing rest mass is evidence of physics beyond the Standard Model of elementary particles. Their absolute mass bears important relevance from particle physics to cosmology. In this work, we report on the search for the effective electron antineutrino mass with the KATRIN experiment. KATRIN performs precision spectroscopy of the tritium $\beta$-decay close to the kinematic endpoint. Based on the first five neutrino-mass measurement campaigns, we derive a best-fit value of $m_\nu^{2} = {-0.14^{+0.13}_{-0.15}}~\mathrm{eV^2}$, resulting in an upper limit of $m_\nu < {0.45}~\mathrm{eV}$ at 90 % confidence level. With six times the statistics of previous data sets, amounting to 36 million electrons collected in 259 measurement days, a substantial reduction of the background level and improved systematic uncertainties, this result tightens KATRIN's previous bound by a factor of almost two., Comment: 61 pages, 20 figures, 2 tables

We study the fixed points of outer-totalistic cellular automata on sparse random regular graphs. These can be seen as constraint satisfaction problems, where each variable must adhere to the same local constraint, which depends solely on its state and the total number of its neighbors in each possible state. Examples of this setting include classical problems such as independent sets or assortative/dissasortative partitions. We analyse the existence and number of fixed points in the large system limit using the cavity method, under both the replica symmetric (RS) and one-step replica symmetry breaking (1RSB) assumption. This method allows us to characterize the structure of the space of solutions, in particular, if the solutions are clustered and whether the clusters contain frozen variables. This last property is conjectured to be linked to the typical algorithmic hardness of the problem. We bring experimental evidence for this claim by studying the performance of the belief-propagation reinforcement algorithm, a message-passing-based solver for these constraint satisfaction problems., Comment: 26 pages, 7 figures

The increasing demand for sustainable energy solutions has driven the integration of digitalized buildings into the power grid, leveraging Internet-of-Things (IoT) technologies to enhance energy efficiency and operational performance. Despite their potential, effectively utilizing IoT point data within deep-learning frameworks presents significant challenges, primarily due to its inherent heterogeneity. This study investigates the diverse dimensions of IoT data heterogeneity in both intra-building and inter-building contexts, examining their implications for predictive modeling. A benchmarking analysis of state-of-the-art time series models highlights their performance on this complex dataset. The results emphasize the critical need for multi-modal data integration, domain-informed modeling, and automated data engineering pipelines. Additionally, the study advocates for collaborative efforts to establish high-quality public datasets, which are essential for advancing intelligent and sustainable energy management systems in digitalized buildings., Comment: 4 figures, 1 tables, 9 pages

In convection-dominated flows, robustness of the spatial discretisation is a key property. While Interior Penalty Galerkin (IPG) methods already proved efficient in the situation of large mesh Peclet numbers, Arbitrary Lagrangian-Eulerian (ALE) methods are able to reduce the convection-dominance by moving the mesh. In this paper, we introduce and analyse a velocity-based moving mesh discontinuous Galerkin method for the solution of the linear advection-diffusion equation. By introducing a smooth parameterized velocity $\tilde{V}$ that separates the flow into a mean flow, also called moving mesh velocity, and a remaining advection field $V-\tilde{V}$, we made a convergence analysis based on the smoothness of the mesh velocity. Furthermore, the reduction of the advection speed improves the stability of an explicit time-stepping and the use of the nonconservative ALE formulation changes the coercivity condition. Finally, by adapting the existing robust error criteria to this moving mesh situation, we derived robust \textit{a posteriori} error criteria that describe the potentially small deviation to the mean flow and include the information of a transition towards $V=\tilde{V}$., Comment: 20 pages, 2 figures, Submitted to SINUM on 16/05/2024, not yet reviewed (15/05/2024)

Sulphur-bearing species are detected in various environments within Galactic star-forming regions and are particularly abundant in the gas phase of outflows and shocks, and photo-dissociation regions. In this work, we aim to investigate the nature of the emission from the most common sulphur-bearing species observable at millimetre wavelengths towards the nuclear starburst of the galaxy NGC 253. We intend to understand which type of regions are probed by sulphur-bearing species and which process(es) dominate(s) the release of sulphur into the gas phase. We used the high-angular resolution (1.6" or 27 pc) observations from the ALCHEMI ALMA Large Program to image several sulphur-bearing species towards the central molecular zone (CMZ) of NGC 253. We performed local thermodynamic equilibrium (LTE) and non-LTE large velocity gradient (LVG) analyses to derive the physical conditions of the gas in which S-bearing species are emitted, and their abundance ratios across the CMZ. Finally, we compared our results with previous ALCHEMI studies and a few selected Galactic environments. We found that not all sulphur-bearing species trace the same type of gas: strong evidence indicates that H2S and part of the emission of OCS, H2CS, and SO, are tracing shocks whilst part of SO and CS emission rather trace the dense molecular gas. For some species, such as CCS and SO2, we could not firmly conclude on their origin of emission. The present analysis indicates that the emission from most sulphur-bearing species throughout the CMZ is likely dominated by shocks associated with ongoing star formation. In the inner part of the CMZ where the presence of super star clusters was previously indicated, we could not distinguish between shocks or thermal evaporation as the main process releasing the S-bearing species., Comment: 44 pages, 20 figures, Accepted for publication in A&A

In clinical settings, we often face the challenge of building prediction models based on small observational data sets. For example, such a data set might be from a medical center in a multi-center study. Differences between centers might be large, thus requiring specific models based on the data set from the target center. Still, we want to borrow information from the external centers, to deal with small sample sizes. There are approaches that either assign weights to each external data set or each external observation. To incorporate information on differences between data sets and observations, we propose an approach that combines both into weights that can be incorporated into a likelihood for fitting regression models. Specifically, we suggest weights at the data set level that incorporate information on how well the models that provide the observation weights distinguish between data sets. Technically, this takes the form of inverse probability weighting. We explore different scenarios where covariates and outcomes differ among data sets, informing our simulation design for method evaluation. The concept of effective sample size is used for understanding the effectiveness of our subgroup modeling approach. We demonstrate our approach through a clinical application, predicting applied radiotherapy doses for cancer patients. Generally, the proposed approach provides improved prediction performance when external data sets are similar. We thus provide a method for quantifying similarity of external data sets to the target data set and use this similarity to include external observations for improving performance in a target data set prediction modeling task with small data., Comment: minor update of affiliations and funding

The Sherrington-Kirkpatrick (SK) model is a prototype of a complex non-convex energy landscape. Dynamical processes evolving on such landscapes and locally aiming to reach minima are generally poorly understood. Here, we study quenches, i.e. dynamics that locally aim to decrease energy. We analyse the energy at convergence for two distinct algorithmic classes, single-spin flip and synchronous dynamics, focusing on greedy and reluctant strategies. We provide precise numerical analysis of the finite size effects and conclude that, perhaps counter-intuitively, the reluctant algorithm is compatible with converging to the ground state energy density, while the greedy strategy is not. Inspired by the single-spin reluctant and greedy algorithms, we investigate two synchronous time algorithms, the sync-greedy and sync-reluctant algorithms. These synchronous processes can be analysed using dynamical mean field theory (DMFT), and a new backtracking version of DMFT. Notably, this is the first time the backtracking DMFT is applied to study dynamical convergence properties in fully connected disordered models. The analysis suggests that the sync-greedy algorithm can also achieve energies compatible with the ground state, and that it undergoes a dynamical phase transition.

Embodied reasoning systems integrate robotic hardware and cognitive processes to perform complex tasks typically in response to a natural language query about a specific physical environment. This usually involves changing the belief about the scene or physically interacting and changing the scene (e.g. 'Sort the objects from lightest to heaviest'). In order to facilitate the development of such systems we introduce a new simulating environment that makes use of MuJoCo physics engine and high-quality renderer Blender to provide realistic visual observations that are also accurate to the physical state of the scene. Together with the simulator we propose a new benchmark composed of 10 classes of multi-step reasoning scenarios that require simultaneous visual and physical measurements. Finally, we develop a new modular Closed Loop Interactive Reasoning (CLIER) approach that takes into account the measurements of non-visual object properties, changes in the scene caused by external disturbances as well as uncertain outcomes of robotic actions. We extensively evaluate our reasoning approach in simulation and in the real world manipulation tasks with a success rate above 76% and 64%, respectively.