Search

The Statistical Finite Element Method (statFEM) offers a Bayesian framework for integrating computational models with observational data, thus providing improved predictions for structural health monitoring and digital twinning. This paper presents an efficient sampling-free statFEM tailored for non-conjugate, non-Gaussian prior probability densities. We assume that constitutive parameters, modeled as weakly stationary random fields, are the primary source of uncertainty and approximate them using Karhunen-Lo\`eve (KL) expansion. The resulting stochastic solution field, i.e., the displacement field, is a non-stationary, non-Gaussian random field, which we approximate via Polynomial Chaos (PC) expansion. The PC coefficients are determined through projection using Smolyak sparse grids. Additionally, we model the measurement noise as a stationary Gaussian random field and the model misspecification as a mean-free, non-stationary Gaussian random field, which is also approximated using KL expansion. The coefficients of the KL expansion are treated as hyperparameters. The PC coefficients of the stochastic posterior displacement field are computed using the Gauss-Markov-K\'alm\'an filter, while the hyperparameters are determined by maximizing the marginal likelihood. We demonstrate the efficiency and convergence of the proposed method through one- and two-dimensional elastostatic problems., Comment: 39 Pages, 20 figures

Patterned nanomagnet arrays (PNAs) have been shown to exhibit a strong geometrically frustrated dipole interaction. Some PNAs have also shown emergent domain wall dynamics. Previous works have demonstrated methods to physically probe these magnetization dynamics of PNAs to realize neuromorphic reservoir systems that exhibit chaotic dynamical behavior and high-dimensional nonlinearity. These PNA reservoir systems from prior works leverage echo state properties and linear/nonlinear short-term memory of component reservoir nodes to map and preserve the dynamical information of the input time-series data into nondelay spatial embeddings. Such mappings enable these PNA reservoir systems to imitate and predict/forecast the input time series data. However, these prior PNA reservoir systems are based solely on the nondelay spatial embeddings obtained at component reservoir nodes. As a result, they require a massive number of component reservoir nodes, or a very large spatial embedding (i.e., high-dimensional spatial embedding) per reservoir node, or both, to achieve acceptable imitation and prediction accuracy. These requirements reduce the practical feasibility of such PNA reservoir systems. To address this shortcoming, we present a mixed delay/nondelay embeddings-based PNA reservoir system. Our system uses a single PNA reservoir node with the ability to obtain a mixture of delay/nondelay embeddings of the dynamical information of the time-series data applied at the input of a single PNA reservoir node. Our analysis shows that when these mixed delay/nondelay embeddings are used to train a perceptron at the output layer, our reservoir system outperforms existing PNA-based reservoir systems for the imitation of NARMA 2, NARMA 5, NARMA 7, and NARMA 10 time series data, and for the short-term and long-term prediction of the Mackey Glass time series data.

Traction force microscopy is a method widely used in biophysics and cell biology to determine forces that biological cells apply to their environment. In the experiment, the cells adhere to a soft elastic substrate, which is then deformed in response to cellular traction forces. The inverse problem consists in computing the traction stress applied by the cell from microscopy measurements of the substrate deformations. In this work, we consider a linear model, in which 3D forces are applied at a 2D interface, called 2.5D traction force microscopy, and a nonlinear pure 2D model, from which we directly obtain a linear pure 2D model. All models lead to a linear resp. nonlinear parameter identification problem for a boundary value problem of elasticity. We analyze the respective forward operators and conclude with some numerical experiments for simulated and experimental data., Comment: 28 pages, 9 figures

In condensed matter systems with the Coulomb interaction playing an important role one expects, besides the on-site (local) Hubbard-type interaction, that also other (non-local) terms depending on the site occupancy, known as correlated or assisted hopping, exist. Even though such terms in quantum dots tunnel coupled to external electrodes may have quite an appreciable amplitude, the interpretation of experiments on these systems -- usually in the linear response regime -- seems not to require their presence. However, since the correlated-hopping term breaks the particle-hole symmetry of the standard Anderson model and modifies all transport characteristics of the system, the detailed knowledge of its influence on measurable characteristics, especially {\em in the non-linear regime}, is a prerequisite for its experimental detection. In this paper, we study the non-linear transport properties of junctions composed of a quantum dot tunnel coupled to external electrodes. We model the system by the single-impurity Anderson Hamiltonian with Hubbard on-site interaction and with a non-local correlated-hopping interaction. Using the previously found general expression for the spin-dependent transport and spectral Green functions as well as general formulae for charge and heat transport, we calculate relevant transport characteristics in the strongly non-linear regime, Comment: 10 pages, 8 figures, 5 pages Suppl. Material; to be published in Ann. Phys. (Berlin)

Fast entangling gate operations are a fundamental prerequisite for quantum simulation and computation. We propose an entangling scheme for arbitrary pairs of ions in a linear crystal, harnessing the high electric polarizability of highly excited Rydberg states. An all-to-all quantum gate connectivity is based on an initialization of a pair of ions to a superposition of ground- and Rydberg-states by laser excitation, followed by the entangling gate operation which relies on a state-dependent frequency shift of collective vibrational modes of the crystal. This gate operation requires applying an electric waveform to trap electrodes. Employing transverse collective modes of oscillation, we reveal order of $\mu s$ operation times within any of the qubit pairs in a small crystal. In our calculation, we are taking into account realistic experimental conditions and feasible electric field ramps. The proposed gate operation is ready to be combined with a scalable processor architecture to reconfigure the qubit register, either by shuttling ions or by dynamically controlling optical tweezer potentials., Comment: 9 pages, 4 figures

Bismuth ferrite (BiFeO3) is one of the rare materials that exhibits multiferroic properties already at room-temperature. Therefore, it offers tremendous potential for future technological applications, such as memory and logic. However, a weak magnetoelectric coupling together with the presence of a noncollinear cycloidal spin order restricts various practical applications of BiFeO3. Therefore, there is a large interest in the search for suitable methods for the modulation of the spin cycloid in BiFeO3. By performing neutron diffraction experiments using a triple-axis instrument we have determined that the spin cycloid can be systematically suppressed by applying a high magnetic field of 10 T in a BiFeO3 thin film of about 100 nm grown on a (110)-oriented SrTiO3 substrate. As predicted by previous theoretical calculations, we observed that the required critical magnetic field to suppress the spin cycloid in a BiFeO3 thin film was lower as compared to the previously reported critical magnetic field for bulk BiFeO3 single crystals. Our experiment reveals that the spin cycloid continuously expands with increasing magnetic field before the complete transformation into a G-type antiferromagnetic spin order. Such tuning of the length of the spin cycloid up to a complete suppression offers new functionalities for future technological applications as in spintronics or magnonics.

We investigate a scheduling problem arising from a material handling and processing problem in a production line of an Austrian company building prefabricated house walls. The addressed problem is a permutation flow shop with blocking constraints in which the machine of at least one stage can process a number operations of two other stages in the system. This situation is usually referred to as multi-task or inter-stage flexibility. The problem is in general NP-hard, but we derive a number of special cases that can be solved in polynomial time. For the general case, we present a variety of heuristic algorithms, with a focus on matheuristics that are grounded in two different mixed-integer linear programming (MIP) formulations of the problem. These matheuristics leverage the strengths of exact optimization techniques while introducing flexibility to address limits on computation time. To assess the performance of the proposed approaches, we conduct an extensive computational study on randomly generated test cases based on real-world instances., Comment: 33 pages, 9 figures, submitted to "Computers & Operations Research" on Nov. 22, 2024

Spin noise spectroscopy is a technique to measure magnetization fluctuations, a subject of increasing relevance in ultrafast spintronics. We investigate numerically the equilibrium spin noise of ferro- and antiferromagnets within an atomistic spin model. The aim is to predict the possible outcomes of ultrafast spin-noise spectroscopy measurements and demonstrate what relevant information can be extracted. Specifically, we show how this method can be used to determine phase transitions, frequencies of magnon modes and correlation times.

This paper presents our approach to scaling the nnU-Net framework for multi-structure segmentation on Cone Beam Computed Tomography (CBCT) images, specifically in the scope of the ToothFairy2 Challenge. We leveraged the nnU-Net ResEnc L model, introducing key modifications to patch size, network topology, and data augmentation strategies to address the unique challenges of dental CBCT imaging. Our method achieved a mean Dice coefficient of 0.9253 and HD95 of 18.472 on the test set, securing a mean rank of 4.6 and with it the first place in the ToothFairy2 challenge. The source code is publicly available, encouraging further research and development in the field., Comment: Fabian Isensee and Yannick Kirchhoff contributed equally

Large ring lasers employed in geodesy and fundamental physics require stability of the perimeter at or below the parts-per-billion level. We present two complementary approaches to actively control the perimeter length of such ring lasers, reaching a relative length stability of $4\times 10^{-10}$. These methods can readily be implemented and bring the stability of heterolithic devices on par with monolithic designs., Comment: 5 pages, 4 figures

Van Hove singularities (vHs) located close to the Fermi level in Kagome superconductors AV3Sb5 (A = K, Rb, Cs) have profound influence on their electronic and transport characteristics. Specifically, magneto-transport and susceptibility measurements on CsV3Sb5 reveal an anomalous temperature dependence of the upper critical field H_c2 (T), characterized by a pronounced upward curvature for both in-plane and c-axis magnetic fields, with zero-temperature H_c2 values of ~6.0 T and ~1.2 T, respectively. Our theoretical analysis, using a newly developed single-band model incorporating vHs and gap anisotropy, suggests that the observed upper critical field behavior is predominantly driven by the anisotropy of the Fermi velocity originating from vHs, instead of multi-band effects or gap anisotropy. Increased electron scattering introduced by proton irradiation defects smears out the vHs, reduces anisotropy, and recovers the conventional H_c2 (T) behavior, corroborating our proposed model., Comment: 31 pages, 11 figures

For decades it is known that Quantum Computers might serve as a tool to solve a very specific kind of problems that have long thought to be incalculable. Some of those problems are of a combinatorial nature, with the quantum advantage arising from the exploding size of a huge decision tree. Although this is of high interest as well, there are more opportunities to make use of the quantum advantage among non-perfect information games with a limited amount of steps within the game. Even though it is not possible to answer the question for the winning move in a specific situation, people are rather interested in what choice gives the best outcome in the long run. This leads us to the search for the highest number of paths within the game's decision tree despite the lack of information and, thus, to a maximum of the payoff-function. We want to illustrate on how Quantum Computers can play a significant role in solving these kind of games, using an example of the most popular German card game Skat. Therefore we use quantum registers to encode the game's information properly and construct the corresponding quantum gates in order to model the game progress and obey the rules. Finally, we use a score operator to project the quantum state onto the winning subspace and therefore evaluate the winning probability for each alternative decision by the player to be made by using quantum algorithms, such as quantum counting of the winning paths to gain a possible advantage in computation speed over classical approaches. Thus, we get a reasonable recommendation of how to act at the table due to the payoff-function maximization. This approach is clearly not doable on a classical computer due to the huge tree-search problem and we discuss peculiarities of the problem that may lead to a quantum advantage when exceeding a certain problem size.

Neural networks can learn spurious correlations in the data, often leading to performance disparity for underrepresented subgroups. Studies have demonstrated that the disparity is amplified when knowledge is distilled from a complex teacher model to a relatively "simple" student model. Prior work has shown that ensemble deep learning methods can improve the performance of the worst-case subgroups; however, it is unclear if this advantage carries over when distilling knowledge from an ensemble of teachers, especially when the teacher models are debiased. This study demonstrates that traditional ensemble knowledge distillation can significantly drop the performance of the worst-case subgroups in the distilled student model even when the teacher models are debiased. To overcome this, we propose Adaptive Group Robust Ensemble Knowledge Distillation (AGRE-KD), a simple ensembling strategy to ensure that the student model receives knowledge beneficial for unknown underrepresented subgroups. Leveraging an additional biased model, our method selectively chooses teachers whose knowledge would better improve the worst-performing subgroups by upweighting the teachers with gradient directions deviating from the biased model. Our experiments on several datasets demonstrate the superiority of the proposed ensemble distillation technique and show that it can even outperform classic model ensembles based on majority voting., Comment: Workshop Algorithmic Fairness through the Lens of Metrics and Evaluation at NeurIPS 2024

Understanding and predicting the structure and evolution of coronal mass ejections (CMEs) in the heliosphere remains one of the most sought-after goals in heliophysics and space weather research. A powerful tool for improving current knowledge and capabilities consists of multi-spacecraft observations of the same event, which take place when two or more spacecraft fortuitously find themselves in the path of a single CME. Multi-probe events can not only supply useful data to evaluate the large-scale of CMEs from 1D in-situ trajectories, but also provide additional constraints and validation opportunities for CME propagation models. In this work, we analyse and simulate the coronal and heliospheric evolution of a slow, streamer-blowout CME that erupted on 23 September 2021 and was encountered in situ by four spacecraft approximately equally distributed in heliocentric distance between 0.4 and 1 au. We employ the Open Solar Physics Rapid Ensemble Information (OSPREI) modelling suite in ensemble mode to predict the CME arrival and structure in a hindcast fashion and to compute the "best-fit" solutions at the different spacecraft individually and together. We find that the spread in the predicted quantities increases with heliocentric distance, suggesting that there may be a maximum (angular and radial) separation between an inner and an outer probe beyond which estimates of the in-situ magnetic field orientation (parameterised by flux rope model geometry) increasingly diverge. We discuss the importance of these exceptional observations and the results of our investigation in the context of advancing our understanding of CME structure and evolution as well as improving space weather forecasts., Comment: 20 pages, 12 figures, 3 tables; accepted for publication in MNRAS

Galaxies at redshift $z\sim 1-2$ display high star formation rates (SFRs) with elevated cold gas fractions and column densities. Simulating a self-regulated ISM in a hydrodynamical, self-consistent context, has proven challenging due to strong outflows triggered by supernova (SN) feedback. At sufficiently high gas column densities, and in the absence of magnetic fields, these outflows prevent a quasi-steady disk from forming at all. To this end, we present GHOSDT, a suite of magneto-hydrodynamical simulations that implement ISM physics at high resolution. We demonstrate the importance of magnetic pressure in the stabilization of gas-rich star-forming disks. We show that a relation between the magnetic field and gas surface density emerges naturally from our simulations. We argue that the magnetic field in the dense, star-forming gas, may be set by the SN-driven turbulent gas motions. When compared to pure hydrodynamical runs, we find that the inclusion of magnetic fields increases the cold gas fraction and reduces the disc scale height, both by up to a factor of $\sim 2$, and reduces the star formation burstiness. In dense ($n>100\;\rm{cm}^{-3}$) gas, we find steady-state magnetic field strengths of 10--40 $\mu$G, comparable to those observed in molecular clouds. Finally, we demonstrate that our simulation framework is consistent with the Ostriker & Kim (2022) Pressure Regulated Feedback Modulated Theory of star formation and stellar Feedback., Comment: Submitted to ApJ

Frustrated spin models may lead to the formation of both classical non-collinear spin structures and unique quantum phases including highly entangled quantum spin liquids. Here, we study the entanglement and spatial quantum correlations in linear spin-wave theory around a classical spin-spiral ground state. We find that the entanglement between pairs of sites is short-ranged, and is completely absent in certain cases. In contrast, the entanglement hidden in multi-site clusters is peaked close to phase transitions and shows an asymptotic behavior modulated by the period of the magnetic structure. These findings motivate further exploring the connection in the entanglement properties of fully quantum and of quasiclassical spin models., Comment: 13 pages, 7 figures

Current interactive segmentation approaches, inspired by the success of META's Segment Anything model, have achieved notable advancements, however, they come with substantial limitations that hinder their practical application in 3D radiological scenarios. These include unrealistic human interaction requirements, such as slice-by-slice operations for 2D models on 3D data, a lack of iterative interactive refinement, and insufficient evaluation experiments. These shortcomings prevent accurate assessment of model performance and lead to inconsistent outcomes across studies. The RadioActive benchmark overcomes these challenges by offering a comprehensive and reproducible evaluation of interactive segmentation methods in realistic, clinically relevant scenarios. It includes diverse datasets, target structures, and interactive segmentation methods, and provides a flexible, extendable codebase that allows seamless integration of new models and prompting strategies. We also introduce advanced prompting techniques to enable 2D models on 3D data by reducing the needed number of interaction steps, enabling a fair comparison. We show that surprisingly the performance of slice-wise prompted approaches can match native 3D methods, despite the domain gap. Our findings challenge the current literature and highlight that models not specifically trained on medical data can outperform the current specialized medical methods. By open-sourcing RadioActive, we invite the research community to integrate their models and prompting techniques, ensuring continuous and transparent evaluation of interactive segmentation models in 3D medical imaging., Comment: Undergoing Peer-Review

Traditional compilers, designed for optimizing low-level code, fall short when dealing with modern, computation-heavy applications like image processing, machine learning, or numerical simulations. Optimizations should understand the primitive operations of the specific application domain and thus happen on that level. Domain-specific languages (DSLs) fulfill these requirements. However, DSL compilers reinvent the wheel over and over again as standard optimizations, code generators, and general infrastructure & boilerplate code must be reimplemented for each DSL compiler. This paper presents MimIR, an extensible, higher-order intermediate representation. At its core, MimIR is a pure type system and, hence, a form of a typed lambda calculus. Developers can declare the signatures of new (domain-specific) operations, called "axioms". An axiom can be the declaration of a function, a type constructor, or any other entity with a possibly polymorphic, polytypic, and/or dependent type. This way, developers can extend MimIR at any low or high level and bundle them in a "plugin". Plugins extend the compiler and take care of optimizing and lowering the plugins' axioms. We show the expressiveness and effectiveness of MimIR in three case studies: Low-level plugins that operate at the same level of abstraction as LLVM, a regular-expression matching plugin, and plugins for linear algebra and automatic differentiation. We show that in all three studies, MimIR produces code that has state-of-the-art performance., Comment: POPL'25

Cell extrusion is an essential mechanism for controlling cell density in epithelial tissues. Another essential element of epithelia is curvature, which is required to achieve complex shapes, like in the lung or intestine. Here we introduce a three-dimensional bubbly vertex model to study the interplay between extrusion and curvature. We find a generic cellular bulging instability at topological defects which is much stronger than for standard vertex models. To clarify the underlying mechanism, we analytically calculate the energy landscape for a single bulging cell in a mean field tissue. We find that extrusion is energetically preferred at topological defects due to a redistribution of Gaussian curvature into the defect. We show that it can be driven both by a decrease in apico-basal tension or by contractile line tensions. We also show that extrusion can be suppressed by luminal pressure and interfacial bending rigidity, allowing for additional control of this important process. Our theory suggests that epithelial curvature naturally leads to cell bulging and extrusion because the interfacial curvature of individual cells at the defects strongly amplifies buckling effected by tissue-scale topological defects in elastic sheets., Comment: 15 pages, 10 figures

We provide a complete characterization of the solvability/impossibility of deterministic stabilizing consensus in any computing model with benign process and communication faults using point-set topology. Relying on the topologies for infinite executions introduced by Nowak, Schmid and Winkler (JACM, 2024) for terminating consensus, we prove that semi-open decision sets and semi-continuous decision functions as introduced by Levin (AMM, 1963) are the appropriate means for this characterization: Unlike the decision functions for terminating consensus, which are continuous, semi-continuous functions do not require the inverse image of an open set to be open and hence allow to map a connected space to a disconnected one. We also show that multi-valued stabilizing consensus with weak and strong validity are equivalent, as is the case for terminating consensus. By applying our results to (variants of) all the known possibilities/impossibilities for stabilizing consensus, we easily provide a topological explanation of these results.

Photonic integrated circuits find ubiquitous use in various technologies, from communication, to computing and sensing, and therefore play a crucial role in the quantum technology counterparts. Several systems are currently under investigation, each showing distinct advantages and drawbacks. For this reason, efforts are made to effectively combine different platforms in order to benefit from their respective strengths. In this work, 3D laser written photonic wire bonds are employed to interface triggered sources of quantum light, based on semiconductor quantum dots embedded into etched microlenses, with low-loss silicon-nitride photonics. Single photons at telecom wavelengths are generated by the In(Ga)As quantum dots which are then funneled into a silicon-nitride chip containing single-mode waveguides and beamsplitters. The second-order correlation function of g(2)(0) = 0.11+/-0.02, measured via the on-chip beamsplitter, clearly demonstrates the transfer of single photons into the silicon-nitride platform. The photonic wire bonds funnel on average 28.6+/-8.8% of the bare microlens emission (NA = 0.6) into the silicon-nitride-based photonic integrated circuit even at cryogenic temperatures. This opens the route for the effective future up-scaling of circuitry complexity based on the use of multiple different platforms.

Cofaces -- simplices that contain a given simplex -- have multiple important uses in generating and using a Vietoris-Rips filtration: both in creating the coboundary matrix for computing persistent cohomology, and for generating the ordered sequence of simplices in the first place. Traditionally, most methods have generated simplices first, and then sorted them in filtration order after the generation step. In this paper, we propose fast algorithms for generating the sequence of simplices by generating cofaces of a given simplex with the same diameter, which by construction produces simplices in filtration order, and for generating additional cofaces in filtration order using sorted neighborhood lists in order to generate coboundaries directly in filtration order.

We introduce a family of boundary conditions and point constraints for conformal immersions that increase the controllability of surfaces defined as minimizers of conformal variational problems. Our free boundary conditions fix the metric on the boundary, up to a global scale, and admit a discretization compatible with discrete conformal equivalence. We also introduce constraints on the conformal scale factor, enforcing rigidity of the geometry in regions of interest, and describe how in the presence of point constraints the conformal class encodes knot points of the spline that can be directly manipulated. To control the tangent planes, we introduce flux constraints balancing the internal material stresses. The collection of these point constraints provide intuitive controls for exploring a subspace of conformal immersions interpolating a fixed set of points in space. We demonstrate the applicability of our framework to geometric modeling, mathematical visualization, and form finding., Comment: to appear in Differential Geom. Appl., 26 pages

Chiral phonons are desirable for applications in spintronics but their generation and control remains a challenge. Here we demonstrate the emergence of truly chiral phonons from selective magnon-phonon coupling in inversion-symmetric magnetic systems. Considering bcc Fe as example, we quantitatively calculate hybridized magnon-phonon quasiparticle states across the entire Brillouin zone utilizing first-principles calculations. Our findings challenge conventional magneto-elastic interpretations and reveal finite zero-point phonon angular momentum and strong anomalous Hall responses linked to finite (spin) Berry curvatures. Our results further establish that the existence of chiral phonons, particularly along high-symmetry directions, is common in many magnetic materials, offering promising avenues for novel spintronic and phononic devices.

How can we test AI performance? This question seems trivial, but it isn't. Standard benchmarks often have problems such as in-distribution and small-size test sets, oversimplified metrics, unfair comparisons, and short-term outcome pressure. As a consequence, good performance on standard benchmarks does not guarantee success in real-world scenarios. To address these problems, we present Touchstone, a large-scale collaborative segmentation benchmark of 9 types of abdominal organs. This benchmark is based on 5,195 training CT scans from 76 hospitals around the world and 5,903 testing CT scans from 11 additional hospitals. This diverse test set enhances the statistical significance of benchmark results and rigorously evaluates AI algorithms across various out-of-distribution scenarios. We invited 14 inventors of 19 AI algorithms to train their algorithms, while our team, as a third party, independently evaluated these algorithms on three test sets. In addition, we also evaluated pre-existing AI frameworks--which, differing from algorithms, are more flexible and can support different algorithms--including MONAI from NVIDIA, nnU-Net from DKFZ, and numerous other open-source frameworks. We are committed to expanding this benchmark to encourage more innovation of AI algorithms for the medical domain., Comment: Accepted to NeurIPS-2024

Topological correctness plays a critical role in many image segmentation tasks, yet most networks are trained using pixel-wise loss functions, such as Dice, neglecting topological accuracy. Existing topology-aware methods often lack robust topological guarantees, are limited to specific use cases, or impose high computational costs. In this work, we propose a novel, graph-based framework for topologically accurate image segmentation that is both computationally efficient and generally applicable. Our method constructs a component graph that fully encodes the topological information of both the prediction and ground truth, allowing us to efficiently identify topologically critical regions and aggregate a loss based on local neighborhood information. Furthermore, we introduce a strict topological metric capturing the homotopy equivalence between the union and intersection of prediction-label pairs. We formally prove the topological guarantees of our approach and empirically validate its effectiveness on binary and multi-class datasets. Our loss demonstrates state-of-the-art performance with up to fivefold faster loss computation compared to persistent homology methods.

Multi-view data from the same source often exhibit correlation. This is mirrored in correlation between the latent spaces of separate variational autoencoders (VAEs) trained on each data-view. A multi-view VAE approach is proposed that incorporates a joint prior with a non-zero correlation structure between the latent spaces of the VAEs. By enforcing such correlation structure, more strongly correlated latent spaces are uncovered. Using conditional distributions to move between these latent spaces, missing views can be imputed and used for downstream analysis. Learning this correlation structure involves maintaining validity of the prior distribution, as well as a successful parameterization that allows end-to-end learning., Comment: Accepted at 'UniReps: 2nd Edition of the Workshop on Unifying Representations in Neural Models', a workshop at NeurIPS 2024

The eROSITA instrument on board Spectrum-Roentgen-Gamma has completed four scans of the X-ray sky, leading to the detection of almost one million X-ray sources in eRASS1 only, including multiple new X-ray binary candidates. We report on analysis of the X-ray binary 1eRASS J085039.9-421151, using a ~55\,ks long NuSTAR observation, following its detection in each eROSITA scan. Analysis of the eROSITA and NuSTAR X-ray spectra in combination with X-shooter data of the optical counterpart provide evidence of an X-ray binary with a red supergiant (RSG) companion, confirming previous results, although we determine a cooler spectral type of M2-3, owing to the presence of TiO bands in the optical and near infrared spectra. The X-ray spectrum is well-described by an absorbed power law with a high energy cutoff typically applied for accreting high mass X-ray binaries. In addition, we detect a strong fluorescent neutral iron line with an equivalent width of ~700\,eV and an absorption edge, the latter indicating strong absorption by a partial covering component. It is unclear if the partial absorber is ionised. There is no significant evidence of a cyclotron resonant scattering feature. We do not detect any pulsations in the NuSTAR lightcurves, possibly on account of a large spin period that goes undetected due to insufficient statistics at low frequencies or potentially large absorption that causes pulsations to be smeared out. Even so, the low persistent luminosity, the spectral parameters observed (photon index, $\Gamma<1.0$), and the minuscule likelihood of detection of RSG-black hole systems, suggest that the compact object is a neutron star., Comment: 17 pages, 11 figures, accepted by A&A

We interpret the coarse symbol and index class of a Callias type Dirac operator $D+\Psi$ on a manifold $M$ as a pairing between the coarse symbol and index classes associated to $D$ and K-theory classes of the coarse corona of $M$ or $M$ itself determined by $\Psi$. Local positivity of $D$ and local invertibility of $\Psi$ are incorporated in terms of support conditions on the $K$-theoretic level., Comment: 106 pages

Deep learning models in medical contexts face challenges like data scarcity, inhomogeneity, and privacy concerns. This study focuses on improving ventricular segmentation in brain MRI images using synthetic data. We employed two latent diffusion models (LDMs): a mask generator trained using 10,000 masks, and a corresponding SPADE image generator optimized using 6,881 scans to create an MRI conditioned on a 3D brain mask. Conditioning the mask generator on ventricular volume in combination with classifier-free guidance enabled the control of the ventricular volume distribution of the generated synthetic images. Next, the performance of the synthetic data was tested using three nnU-Net segmentation models trained on a real, augmented and entirely synthetic data, respectively. The resulting models were tested on a completely independent hold-out dataset of patients with enlarged ventricles, with manual delineation of the ventricles used as ground truth. The model trained on real data showed a mean absolute error (MAE) of 9.09 \pm 12.18 mL in predicted ventricular volume, while the models trained on synthetic and augmented data showed MAEs of 7.52 \pm 4.81 mL and 6.23 \pm 4.33 mL, respectively. Both the synthetic and augmented model also outperformed the state-of-the-art model SynthSeg, which due to limited performance in cases of large ventricular volumes, showed an MAE of 7.73 \pm 12.12 mL with a factor of 3 higher standard deviation. The model trained on augmented data showed the highest Dice score of 0.892 \pm 0.05, slightly outperforming SynthSeg and on par with the model trained on real data. The synthetic model performed similar to SynthSeg. In summary, we provide evidence that guided synthesis of labeled brain MRI data using LDMs improves the segmentation of enlarged ventricles and outperforms existing state-of-the-art segmentation models.

We explore the gravitational-wave emission from head-on collisions of equal-mass solitonic boson-star binaries from simulations spanning a two-dimensional parameter space, consisting of the central scalar-field amplitude of the stars and the solitonic potential parameter. We report the gravitational-wave energies emitted by boson-star binaries which, due to their combination of moderately high compactness with significant deformability, we often find to be louder by up to an order of magnitude than analogous black-hole collisions. The dependence of the radiated energy on the boson-star parameters exhibits striking needle-sharp features and discontinuous jumps to the value emitted by black-hole binaries. We explain these features in terms of the solitonic potential and the stability properties of the respective individual stars., Comment: 17 pages, 10 figures

We introduce new high-resolution galaxy simulations accelerated by a surrogate model that reduces the computation cost by approximately 75 percent. Massive stars with a Zero Age Main Sequence mass of about 8 solar masses and above explode as core-collapse supernovae (CCSNe), which play a critical role in galaxy formation. The energy released by CCSNe is essential for regulating star formation and driving feedback processes in the interstellar medium (ISM). However, the short integration timesteps required for SNe feedback present significant bottlenecks in star-by-star galaxy simulations that aim to capture individual stellar dynamics and the inhomogeneous shell expansion of SNe within the turbulent ISM. Our new framework combines direct numerical simulations and surrogate modeling, including machine learning and Gibbs sampling. The star formation history and the time evolution of outflow rates in the galaxy match those obtained from resolved direct numerical simulations. Our new approach achieves high-resolution fidelity while reducing computational costs, effectively bridging the physical scale gap and enabling multi-scale simulations., Comment: 20 pages, 14 figures, 3 tables, submitted to ApJ

Self-Supervised Learning (SSL) presents an exciting opportunity to unlock the potential of vast, untapped clinical datasets, for various downstream applications that suffer from the scarcity of labeled data. While SSL has revolutionized fields like natural language processing and computer vision, its adoption in 3D medical image computing has been limited by three key pitfalls: Small pre-training dataset sizes, architectures inadequate for 3D medical image analysis, and insufficient evaluation practices. In this paper, we address these issues by i) leveraging a large-scale dataset of 39k 3D brain MRI volumes and ii) using a Residual Encoder U-Net architecture within the state-of-the-art nnU-Net framework. iii) A robust development framework, incorporating 5 development and 8 testing brain MRI segmentation datasets, allowed performance-driven design decisions to optimize the simple concept of Masked Auto Encoders (MAEs) for 3D CNNs. The resulting model not only surpasses previous SSL methods but also outperforms the strong nnU-Net baseline by an average of approximately 3 Dice points setting a new state-of-the-art. Our code and models are made available here., Comment: Arxiv Preprint. Revised and under review

What representation do deep neural networks learn? How similar are images to each other for neural networks? Despite the overwhelming success of deep learning methods key questions about their internal workings still remain largely unanswered, due to their internal high dimensionality and complexity. To address this, one approach is to measure the similarity of activation responses to various inputs. Representational Similarity Matrices (RSMs) distill this similarity into scalar values for each input pair. These matrices encapsulate the entire similarity structure of a system, indicating which input leads to similar responses. While the similarity between images is ambiguous, we argue that the spatial location of semantic objects does neither influence human perception nor deep learning classifiers. Thus this should be reflected in the definition of similarity between image responses for computer vision systems. Revisiting the established similarity calculations for RSMs we expose their sensitivity to spatial alignment. In this paper, we propose to solve this through semantic RSMs, which are invariant to spatial permutation. We measure semantic similarity between input responses by formulating it as a set-matching problem. Further, we quantify the superiority of semantic RSMs over spatio-semantic RSMs through image retrieval and by comparing the similarity between representations to the similarity between predicted class probabilities., Comment: Accepted at NeurIPS2024

We present the results of Phase I of an ongoing review of Monte Carlo tools relevant for low-energy hadronic cross sections. This includes a detailed comparison of Monte Carlo codes for electron-positron scattering into a muon pair, pion pair, and electron pair, for scan and radiative-return experiments. After discussing the various approaches that are used and effects that are included, we show differential cross sections obtained with AfkQed, BabaYaga@NLO, KKMC, MCGPJ, McMule, Phokhara, and Sherpa, for scenarios that are inspired by experiments providing input for the dispersive evaluation of the hadronic vacuum polarisation., Comment: RadioMonteCarLow 2 Working Group report Phase I, 67 pages, 34 figures; references added and other minor modifications

We consider a maximum entropy edge weight model for shortest path networks that allows for negative weights. Given a graph $G$ and possible weights $\mathcal{W}$ typically consisting of positive and negative values, the model selects edge weights $w \in \mathcal{W}^m$ uniformly at random from all weights that do not introduce a negative cycle. We propose an MCMC process and show that (i) it converges to the required distribution and (ii) that the mixing time on the cycle graph is polynomial. We then engineer an implementation of the process using a dynamic version of Johnson's algorithm in connection with a bidirectional Dijkstra search. We empirically study the performance characteristics of an implementation of the novel sampling algorithm as well as the output produced by the model.

We introduce motion graph, a novel approach to the video prediction problem, which predicts future video frames from limited past data. The motion graph transforms patches of video frames into interconnected graph nodes, to comprehensively describe the spatial-temporal relationships among them. This representation overcomes the limitations of existing motion representations such as image differences, optical flow, and motion matrix that either fall short in capturing complex motion patterns or suffer from excessive memory consumption. We further present a video prediction pipeline empowered by motion graph, exhibiting substantial performance improvements and cost reductions. Experiments on various datasets, including UCF Sports, KITTI and Cityscapes, highlight the strong representative ability of motion graph. Especially on UCF Sports, our method matches and outperforms the SOTA methods with a significant reduction in model size by 78% and a substantial decrease in GPU memory utilization by 47%., Comment: Accepted by NeurIPS 2024, 19 pages, 12 figures

Natural light-harvesting antenna complexes efficiently capture solar energy using chlorophyll, i.e., magnesium porphyrin pigments, embedded in a protein matrix. Inspired by this natural configuration, artificial clay-porphyrin antenna structures have been experimentally synthesized and have demonstrated remarkable excitation energy transfer properties. The study presents the computational design and simulation of a synthetic light-harvesting system that emulates natural mechanisms by arranging cationic free-base porphyrin molecules on an anionic clay surface. We investigated the transfer of excitation energy among the porphyrin dyes using a multiscale quantum mechanics/molecular mechanics (QM/MM) approach based on the semi-empirical density functional-based tight-binding (DFTB) theory for the ground state dynamics. To improve the accuracy of our results, we incorporated an innovative multifidelity machine learning (MFML) approach, which allows the prediction of excitation energies at the numerically demanding time-dependent density functional theory level with the Def2-SVP basis set. This approach was applied to an extensive dataset of 640K geometries for the 90-atom porphyrin structures, facilitating a thorough analysis of the excitation energy diffusion among the porphyrin molecules adsorbed to the clay surface. The insights gained from this study, inspired by natural light-harvesting complexes, demonstrate the potential of porphyrin-clay systems as effective energy transfer systems.

Machine learning interatomic potentials (MLIPs) have seen significant advances as efficient replacement of expensive quantum chemical calculations. Uncertainty estimations for MLIPs are crucial to quantify the additional model error they introduce and to leverage this information in active learning strategies. MLIPs that are based on Gaussian process regression provide a standard deviation as a possible uncertainty measure. An alternative approach are ensemble-based uncertainties. Although these uncertainty measures have been applied to active learning, it has rarely been studied how they correlate with the error, and it is not always clear whether active learning actually outperforms random sampling strategies. We consider GPR models with Coulomb and SOAP representations as inputs to predict potential energy surfaces and excitation energies of molecules. We evaluate, how the GPR variance and ensemble-based uncertainties relate to the error and whether model performance improves by selecting the most uncertain samples from a fixed configuration space. For the ensemble based uncertainty estimations, we find that they often do not provide any information about the error. For the GPR standard deviation, we find that often predictions with an increasing standard deviation also have an increasing systematical bias, which is not captured by the uncertainty. In these cases, selecting training samples with the highest uncertainty leads to a model with a worse test error compared to random sampling. We conclude that confidence intervals, which are derived from the predictive standard deviation, can be highly overconfident. Selecting samples with high GPR standard deviation leads to a model that overemphasizes the borders of the configuration space represented in the fixed dataset. This may result in worse performance in more densely sampled areas but better generalization for extrapolation tasks.

Offloading computationally expensive algorithms to the edge or even cloud offers an attractive option to tackle limitations regarding on-board computational and energy resources of robotic systems. In cloud-native applications deployed with the container management system Kubernetes (K8s), one key problem is ensuring resilience against various types of failures. However, complex robotic systems interacting with the physical world pose a very specific set of challenges and requirements that are not yet covered by failure mitigation approaches from the cloud-native domain. In this paper, we therefore propose a novel approach for robotic system monitoring and stateful, reactive failure mitigation for distributed robotic systems deployed using Kubernetes (K8s) and the Robot Operating System (ROS2). By employing the generic substrate of Behaviour Trees, our approach can be applied to any robotic workload and supports arbitrarily complex monitoring and failure mitigation strategies. We demonstrate the effectiveness and application-agnosticism of our approach on two example applications, namely Autonomous Mobile Robot (AMR) navigation and robotic manipulation in a simulated environment.

Pair-production and single-production of $W$ bosons provide many opportunities to look for new physics via precision measurements, for instance via scrutinising the involved triple-gauge vertices or by measuring CKM matrix elements in an environment very complementary to $B$ hadron decays. This contribution presents the ongoing work based on full simulation of the ILD concept, exploiting the O($10^8$) $W$ bosons produced during the $250$\,GeV stage of the ILC, as well as the CLD concept proposed for the FCC-ee. The projections, which also contribute to two focus topics of the current ECFA study on future Higgs/Top/Electroweak factories, promise improvements of the measurement precision of TGCs and the CKM matrix, respectively, of one to two orders of magnitude with respects to LEP., Comment: Contribution to proceedings of LCWS 2024, 6 pages

The aggressiveness of prostate cancer, the most common cancer in men worldwide, is primarily assessed based on histopathological data using the Gleason scoring system. While artificial intelligence (AI) has shown promise in accurately predicting Gleason scores, these predictions often lack inherent explainability, potentially leading to distrust in human-machine interactions. To address this issue, we introduce a novel dataset of 1,015 tissue microarray core images, annotated by an international group of 54 pathologists. The annotations provide detailed localized pattern descriptions for Gleason grading in line with international guidelines. Utilizing this dataset, we develop an inherently explainable AI system based on a U-Net architecture that provides predictions leveraging pathologists' terminology. This approach circumvents post-hoc explainability methods while maintaining or exceeding the performance of methods trained directly for Gleason pattern segmentation (Dice score: 0.713 $\pm$ 0.003 trained on explanations vs. 0.691 $\pm$ 0.010 trained on Gleason patterns). By employing soft labels during training, we capture the intrinsic uncertainty in the data, yielding strong results in Gleason pattern segmentation even in the context of high interobserver variability. With the release of this dataset, we aim to encourage further research into segmentation in medical tasks with high levels of subjectivity and to advance the understanding of pathologists' reasoning processes., Comment: 58 pages, 15 figures (incl. supplementary)