arXiv subjects

Rolf Krause

Publications and source records attributed to Rolf Krause.

At least 19 recordsLinked to original sources

arXiv2026

Protect the Brain When Treating the Heart: Feasibility of 2.5D U-Net for Real-Time Gaseous Microemboli Detection

Gaseous microemboli (GME) represent a common complication of cardiac structural interventions across both surgical and transcatheter approaches. Intraoperative transesophageal echocardiography (TEE) represents a convenient methodology to monitor and visualize the presence of circulating GME. However, their detection and quantification are far from trivial due to operator-dependent view, high velocity, and objects with similar structure in the background. Here, we propose a feasibility study based on a 2.5D U-Net architecture to detect GME in space-time connected data. We applied and tested such an architecture on a pilot dataset of eight TEE recordings ($60$ fps, $600\times 800$ pixels) from eight different patients undergoing cardiac surgery, resulting in improved detection of moving GMEs against the background with respect to classical spot detection algorithms and 2D U-Net, yet retaining real-time execution speed with respect to more complex deep-learning architectures. Under leave-one-patient-out cross-validation, the selected model achieved strong detection performance under a three-pixel radius-tolerant grace-zone evaluation, with a precision of 92.55\% and recall of 80.54\%, corresponding to radius-tolerant Intersection over Union (IoU) and Dice coefficients of 73.95\% and 84.13\%, respectively. Complementarily, strict pixel-based segmentation metrics were also computed, yielding an IoU of 41.74\% and a Dice coefficient of 57.98\%. The selected model achieved an average inference time of $0.12 s$ per batch on the tested hardware. To assess specificity on unseen data, we additionally evaluated the model on an external GME-negative TEE dataset, where it produced predominantly empty or near-empty masks, indicating a low rate of spurious detections. These results support the technical feasibility of real-time GME segmentation.

cs.LG↗

arXiv2026

Multilevel Preconditioning Strategies for Convex Optimization Methods in Image Deblurring

Proximal gradient methods are widely used in imaging, and their speed of convergence can be accelerated by incorporating variable metrics and/or extrapolation steps. Recent works have shown that preconditioning strategies can significantly enhance this acceleration, in particular, for image deblurring problems. In parallel, a multilevel framework has been introduced to speed up inertial and inexact forward-backward schemes for image restoration problems. In this paper, we combine preconditioning and multilevel strategies to design a robust and consistent acceleration framework for both standard and inexact forward-backward schemes applied to regularized convex optimization problems. Numerical experiments in image deblurring confirm that our approach yields a substantial improvement in convergence speed compared to standard methods.

math.NA↗

arXiv2026

A Non-Monotone Preconditioned Trust-Region Method for Neural Network Training

Training deep neural networks at scale can benefit from domain decomposition, where the network is split into subdomains trained in parallel and coupled by a global trust-region mechanism. Building on the Additively Preconditioned Trust-Region Strategy (APTS), we propose a non-monotone variant with a nonlinear additive Schwarz preconditioner that combines parallel subdomain corrections with global coarse-space directions. A windowed acceptance criterion allows controlled objective increases, avoiding needless rejection of effective coarse steps. The resulting non-monotone APTS (NAPTS) preserves accuracy while reducing CPU time by 30\% and cutting rejected steps to one third of those in APTS.

math.OC↗

arXiv2026

Multi-Preconditioned LBFGS for Training Finite-Basis PINNs

A multi-preconditioned LBFGS (MP-LBFGS) algorithm is introduced for training finite-basis physics-informed neural networks (FBPINNs). The algorithm is motivated by the nonlinear additive Schwarz method and exploits the domain-decomposition-inspired additive architecture of FBPINNs, in which local neural networks are defined on subdomains, thereby localizing the network representation. Parallel, subdomain-local quasi-Newton corrections are then constructed on the corresponding local parts of the architecture. A key feature is a novel nonlinear multi-preconditioning mechanism, in which subdomain corrections are optimally combined through the solution of a low-dimensional subspace minimization problem. Numerical experiments indicate that MP-LBFGS can improve convergence speed, as well as model accuracy over standard LBFGS while incurring lower communication overhead.

math.NA↗

arXiv2026

Structure-preserving preconditioning of discrete space-fractional diffusion equations with variable coefficient and θ-Method

This paper studies the spectral properties of large matrices and the preconditioning of linear systems, arising from the finite difference discretization of a time-dependent space-fractional diffusion equation with a variable coefficient $a(x)$ defined on $Ω\subset \mathbb{R}^d$, $d=1,2$. The model involves a one-sided Riemann-Liouville fractional derivative multiplied by the function $a(x)$, discretized by the shifted Gr"unwald formula in space and the $θ$-method in time. The resulting all-at-once linear systems exhibit a $(d+1)$-level Toeplitz-like matrix structure, with $d=1,2$ denoting the space dimension, while the additional level is due to the time variable. A preconditioning strategy is developed based on the structural properties of the discretized operator. Using the generalized locally Toeplitz (GLT) theory, we analyze the spectral distribution of the unpreconditioned and preconditioned matrix sequences. The main novelty is that the analysis fully covers the case where the variable coefficient $a$ is nonconstant. Numerical results are provided to support the GLT based theoretical findings, and some possible extensions are briefly discussed.

math.NA↗

arXiv2026

3D radiative transfer modeling of scattering polarization with partial frequency redistribution I. Verification and disk-center results for the solar Ca I 4227 Å line

Several strong solar resonance lines show observable linear scattering polarization signals, holding a great potential for investigating the magnetism of the outer solar atmosphere. Accurately modeling these signals requires solving the radiative transfer (RT) problem for polarized radiation in comprehensive 3D models of the solar atmosphere, in non-local thermodynamic equilibrium, accounting for partial frequency redistribution (PRD) effects. This problem has so far been computationally inaccessible. We present the first scientific application of TRIP, a novel software for the massively parallel solution of the 3D non-LTE RT problem for polarized radiation, including scattering polarization and PRD. We aim to verify the code and explore the combined action of PRD and the 3D structure of the solar atmosphere on scattering polarization. We run TRIP to synthesize the Stokes profiles of the Ca I line at 4227 Å in a 3D model of the solar atmosphere extracted from a radiation magneto-hydrodynamic simulation. We efficiently solve the resulting large-scale problem, with up to $4 \times 10^{10}$ degrees of freedom, with a state-of-the-art preconditioned Krylov method, using up to 20 thousand parallel CPUs. After including verification tests, we find that the joint impact of PRD effects and the detailed 3D structure of the atmospheric model produce disk-center scattering polarization signals in the line wings. These signals are sensitive to the magnetic field, via magneto-optical effects, and to bulk velocity gradients. We also show that the CRD approximation underestimates the amplitude of disk-center line-core signals. This achievement represents a crucial step forward for diagnosing the magnetism of the solar chromosphere and transition region through the quantitative comparisons of synthetic and observational data.

astro-ph.SR↗

arXiv2026

Ensured Energy: How a Serious Game can Reach and Engage Diverse Societal Groups in Swiss Energy Transition

In support of Switzerland's energy and climate strategy for 2050, researchers investigate scenarios for the transition of energy systems towards a higher share of renewables, assessing their social, environmental and economic impact. Their results guide stakeholders and policy makers in designing resilient and sustainable systems. A crucial condition to successfully implement these solution in the real-world is that the population supports these transitions. Social scientists have identified the high complexity of energy systems and energy policy as one reason why popular support for the implementation of the energy transition is often limited. This paper proposes serious gaming as a novel approach to inform and sensitize a broader public for the transition of the energy system in Switzerland. We motivate and describe the design of an online game in which players experience an accurate simulation of current and future energy provision and manage transition towards a sustainable future. We present the embedding of this serious game into a large-scale population survey and report findings on player characteristics and engagement. We show that a serious game can successfully attract participants from diverse societal groups and highlight the challenge of balancing complexity and entertainment.

cs.CE↗

arXiv2026

Layer-Parallel Training for Transformers

We present a new training methodology for transformers using a multilevel, layer-parallel approach. Through a neural ODE formulation of transformers, our application of a multilevel parallel-in-time algorithm for the forward and backpropagation phases of training achieves parallel acceleration over the layer dimension. This dramatically enhances parallel scalability as the network depth increases, which is particularly useful for increasingly large foundational models. However, achieving this introduces errors that cause systematic bias in the gradients, which in turn reduces convergence when closer to the minima. We develop an algorithm to detect this critical transition and either switch to serial training or systematically increase the accuracy of layer-parallel training. Results, including BERT, GPT2, ViT, and machine translation architectures, demonstrate parallel-acceleration as well as accuracy commensurate with serial pre-training while fine-tuning is unaffected.

cs.LG↗

arXiv2026

Cell Behavior Video Classification Challenge, a benchmark for computer vision methods in time-lapse microscopy

The classification of microscopy videos capturing complex cellular behaviors is crucial for understanding and quantifying the dynamics of biological processes over time. However, it remains a frontier in computer vision, requiring approaches that effectively model the shape and motion of objects without rigid boundaries, extract hierarchical spatiotemporal features from entire image sequences rather than static frames, and account for multiple objects within the field of view. To this end, we organized the Cell Behavior Video Classification Challenge (CBVCC), benchmarking 35 methods based on three approaches: classification of tracking-derived features, end-to-end deep learning architectures to directly learn spatiotemporal features from the entire video sequence without explicit cell tracking, or ensembling tracking-derived with image-derived features. We discuss the results achieved by the participants and compare the potential and limitations of each approach, serving as a basis to foster the development of computer vision methods for studying cellular dynamics.

eess.IV↗

arXiv2025

An Additively Preconditioned Trust Region Strategy for Machine Learning

Modern machine learning, especially the training of deep neural networks, depends on solving large-scale, highly nonconvex optimization problems, whose objective function exhibit a rough landscape. Motivated by the success of parallel preconditioners in the context of Krylov methods for large scale linear systems, we introduce a novel nonlinearly preconditioned Trust-Region method that makes use of an additive Schwarz correction at each minimization step, thereby accelerating convergence. More precisely, we propose a variant of the Additively Preconditioned Trust-Region Strategy (APTS), which combines a right-preconditioned additive Schwarz framework with a classical Trust-Region algorithm. By decomposing the parameter space into sub-domains, APTS solves local non-linear sub-problems in parallel and assembles their corrections additively. The resulting method not only shows fast convergence; due to the underlying Trust-Region strategy, it furthermore largely obviates the need for hyperparameter tuning.

math.NA↗

arXiv2025

Adaptive Multidimensional Quadrature on Multi-GPU Systems

We introduce a distributed adaptive quadrature method that formulates multidimensional integration as a hierarchical domain decomposition problem on multi-GPU architectures. The integration domain is recursively partitioned into subdomains whose refinement is guided by local error estimators. Each subdomain evolves independently on a GPU, which exposes a significant load imbalance as the adaptive process progresses. To address this challenge, we introduce a decentralised load redistribution schemes based on a cyclic round-robin policy. This strategy dynamically rebalance subdomains across devices through non-blocking, CUDA-aware MPI communication that overlaps with computation. The proposed strategy has two main advantages compared to a state-of-the-art GPU-tailored package: higher efficiency in high dimensions; and improved robustness w.r.t the integrand regularity and the target accuracy.

cs.DC↗

arXiv2025

Trust-Region Methods with Low-Fidelity Objective Models

We introduce two multifidelity trust-region methods based on the Magical Trust Region (MTR) framework. MTR augments the classical trust-region step with a secondary, informative direction. In our approaches, the secondary ``magical'' directions are determined by solving coarse trust-region subproblems based on low-fidelity objective models. The first proposed method, Sketched Trust-Region (STR), constructs this secondary direction using a sketched matrix to reduce the dimensionality of the trust-region subproblem. The second method, SVD Trust-Region (SVDTR), defines the magical direction via a truncated singular value decomposition of the dataset, capturing the leading directions of variability. Several numerical examples illustrate the potential gain in efficiency.

math.NA↗

arXiv2025

A numerical approach for modelling the polarisation signals of strong resonance lines with partial frequency redistribution. Numerical applications to two-term atoms and plane-parallel atmospheres

Aims. The main goal of this paper is to present an accurate and efficient numerical strategy for solving the radiative transfer problem for polarised radiation in strong resonance lines forming out of local thermodynamic equilibrium, taking angle-dependent (AD) partial frequency redistribution (PRD) effects and J-state interference into account. We consider the polarisation produced both by the Zeeman effect and by the scattering of anisotropic radiation, along with its sensitivity to the Hanle and magneto-optical effects. Methods. We introduce a formalism that allows treating both a two-level and a two-term atom in the presence of arbitrary magnetic and bulk velocity fields. The problem is formulated by treating the population of the lower level/term as a fixed input parameter. This approach makes the problem linear with respect to the radiation field, enabling the application of efficient matrix-free preconditioned iterative methods for its solution. Additionally, the computation of the scattering emissivity in the comoving frame, together with a careful choice of the angular and spectral quadrature nodes, allow us to speed up the calculations by reducing the number of evaluations of the redistribution functions. Results. The proposed solution strategy is applied to synthesise the Stokes profiles of the Mg ii h&k doublet and the H i Ly-α line in 1D semi-empirical models. The results demonstrate that the method is both fast and accurate. A comparison with calculations from HanleRT-TIC displays an overall good agreement, thereby validating our solution strategy. Moreover, for the wavelength-integrated polarisation profiles of the H i Ly-α line, we find an excellent agreement between the results obtained including PRD effects in their general AD description and those obtained considering the angle-averaged simplifying approximation.

astro-ph.SR↗

arXiv2025

A collision model for very flexible Cosserat rods and immersed-boundary fluid-structure coupling

The paper presents a constraint-based collision model for Cosserat rods, able to handle dynamic or static contact between a large number of highly flexible structures. The model provides the required collision impulses prior to updating the solution of the rods, with the impulses accounted for as external loads. The procedure avoids the need to modify the structure solver itself and circumvents any iteration between the collision model and the solver for the Cosserat rods, maintaining the efficiency of any chosen Cosserat solver. The collision model is adopted from Tschisgale et al. (Arch. Appl. Mech., 2019) and extended towards higher stability, which is found necessary in the case of very flexible rods. Furthermore, the model is supplemented with additional terms that arise when the colliding rods are immersed in a fluid. The latter is accounted for by an immersed-boundary method. A large number of tests are conducted to demonstrate the functionality of the final model.

cond-mat.soft↗

arXiv2025

Integrating Additive Multigrid with Multipreconditioned Conjugate Gradient Method

Due to its optimal complexity, the multigrid (MG) method is one of the most popular approaches for solving large-scale linear systems arising from the discretization of partial differential equations. However, the parallel implementation of standard MG methods, which are inherently multiplicative, suffers from increasing communication complexity. In such cases, the additive variants of MG methods provide a good alternative due to their inherently parallel nature, although they exhibit slower convergence. This work combines the additive multigrid method with the multipreconditioned conjugate gradient (MPCG) method. In the proposed approach, the MPCG method employs the corrections from the different levels of the MG hierarchy as separate preconditioned search directions. In this approach, the MPCG method updates the current iterate by using the linear combination of the preconditioned search directions, where the optimal coefficients for the linear combination are computed by exploiting the energy norm minimization of the CG method. The idea behind our approach is to combine the $A$-conjugacy of the search directions of the MPCG method and the quasi $H_1$-orthogonality of the corrections from the MG hierarchy. In the numerical section, we study the performance of the proposed method compared to the standard additive and multiplicative MG methods used as preconditioners for the CG method.

math.NA↗

arXiv2025

Data-Parallel Neural Network Training via Nonlinearly Preconditioned Trust-Region Method

Parallel training methods are increasingly relevant in machine learning (ML) due to the continuing growth in model and dataset sizes. We propose a variant of the Additively Preconditioned Trust-Region Strategy (APTS) for training deep neural networks (DNNs). The proposed APTS method utilizes a data-parallel approach to construct a nonlinear preconditioner employed in the nonlinear optimization strategy. In contrast to the common employment of Stochastic Gradient Descent (SGD) and Adaptive Moment Estimation (Adam), which are both variants of gradient descent (GD) algorithms, the APTS method implicitly adjusts the step sizes in each iteration, thereby removing the need for costly hyperparameter tuning. We demonstrate the performance of the proposed APTS variant using the MNIST and CIFAR-10 datasets. The results obtained indicate that the APTS variant proposed here achieves comparable validation accuracy to SGD and Adam, all while allowing for parallel training and obviating the need for expensive hyperparameter tuning.

cs.LG↗

arXiv2024

An eikonal model with re-excitability for fast simulations in cardiac electrophysiology

Precision cardiology based on cardiac digital twins requires accurate simulations of cardiac arrhythmias. However, detailed models, such as the monodomain model, are computationally costly and have limited applicability in practice. Thus, it desirable to have fast models that can still represent the main physiological features presented during cardiac arrhythmias. The eikonal model is an approximation of the monodomain model that is widely used to describe the arrival times of the electrical wave. However, the standard eikonal model does not generalize to the complex re-entrant dynamics that characterize the cardiac arrhythmias. In this work, we propose an eikonal model that includes the tissue re-excitability, which allows to describe re-entries. The re-excitability properties are inferred from the monodomain model. Our eikonal model also handles the tissue anisotropy and heterogeneity. We compare the eikonal model to the monodomain model in various numerical experiments in the atria and the ventricles. The eikonal model is qualitatively accurate in the simulation of re-entries and can be potentially ran in real-time, opening the door to its clinical applicability.

cs.CE↗

arXiv2024

Symbol-based multilevel block $τ$ preconditioners for multilevel block Toeplitz systems: GLT-based analysis and applications

In recent years, there has been a renewed interest in preconditioning for multilevel Toeplitz systems, a research field that has been extensively explored over the past several decades. This work introduces novel preconditioning strategies using multilevel $τ$ matrices for both symmetric and nonsymmetric multilevel Toeplitz systems. Our proposals constitute a general framework, as they are constructed solely based on the generating function of the multilevel Toeplitz coefficient matrix, when it can be defined. We begin with nonsymmetric systems, where we employ a symmetrization technique by permuting the coefficient matrix to produce a real symmetric multilevel Hankel structure. We propose a multilevel $τ$ preconditioner tailored to the symmetrized system and prove that the eigenvalues of the preconditioned matrix sequence cluster at $\pm 1$, leading to rapid convergence when using the preconditioned minimal residual method. The high effectiveness of this approach is demonstrated through its application in solving space fractional diffusion equations. Next, for symmetric systems we introduce another multilevel $τ$ preconditioner and show that the preconditioned conjugate gradient method can achieve an optimal convergence rate, namely a rate that is independent of the matrix size, when employed for a class of ill-conditioned multilevel Toeplitz systems. Numerical examples are provided to critically assess the effectiveness of our proposed preconditioners compared to several leading existing preconditioned solvers, highlighting their superior performance.

math.NA↗