arXiv subjects

Markus Götz

Publications and source records attributed to Markus Götz.

At least 19 recordsLinked to original sources

arXiv2026

4D Parallelism Unlocks Exascale Bayesian Neural Networks for High-Fidelity Atmospheric Modeling

We present BEAST, the first-ever Bayesian Swin Transformer for atmospheric forecasting on 0.25$^\circ$ global resolution able to accurately quantify both aleatoric and epistemic uncertainty. To overcome the associated computational bottlenecks, we devise an orthogonal 4D-parallelization scheme that introduces a unique domain-tensor-parallelism strategy and a novel uncertainty parallel method, enabling us to fully leverage GPU capacity and efficiently scale model training. For a 2.4-billion-parameter model, we achieve a peak performance of 3.96 EFLOP/s on 20,480 NVIDIA GH200 GPUs on the JUPITER supercomputer. We train BEAST as a 700-million-parameter model with 96 random weight samples on 384 nodes on 40 years of data for nearly one million gradient updates. This model achieves predictive skill scores competitive with state-of-the-art probabilistic atmospheric AI models and numerical models, and can predict extreme events with exceptional skill, while generating large ensembles 3 to 4 times faster than the current-best AI model. Our contribution unlocks the potential of high-fidelity uncertainty quantification in atmospheric AI models, heralding a new era for AI-based models in climate and Earth system sciences.

physics.ao-ph↗

arXiv2026

Sampling Parallelism for Fast and Efficient Bayesian Learning

Machine learning models, and deep neural networks in particular, are increasingly deployed in risk-sensitive domains such as healthcare, environmental forecasting, and finance, where reliable quantification of predictive uncertainty is essential. However, many uncertainty quantification (UQ) methods remain difficult to apply due to their substantial computational cost. Sampling-based Bayesian learning approaches, such as Bayesian neural networks (BNNs), are particularly expensive since drawing and evaluating multiple parameter samples rapidly exhausts memory and compute resources. These constraints have limited the accessibility and exploration of Bayesian techniques thus far. To address these challenges, we introduce sampling parallelism, a simple yet powerful parallelization strategy that targets the primary bottleneck of sampling-based Bayesian learning: the samples themselves. By distributing sample evaluations across multiple GPUs, our method reduces memory pressure and training time without requiring architectural changes or extensive hyperparameter tuning. We detail the methodology and evaluate its performance on a few example tasks and architectures, comparing against distributed data parallelism (DDP) as a baseline. We further demonstrate that sampling parallelism is complementary to existing strategies by implementing a hybrid approach that combines sample and data parallelism. Our experiments show near-perfect scaling when the sample number is scaled proportionally to the computational resources, confirming that sample evaluations parallelize cleanly. Although DDP achieves better raw speedups under scaling with constant workload, sampling parallelism has a notable advantage: by applying independent stochastic augmentations to the same batch on each GPU, it increases augmentation diversity and thus reduces the number of epochs required for convergence.

cs.LG↗

arXiv2026

Inverse Design of Optical Multilayer Thin Films using Robust Masked Diffusion Models

Inverse design of optical multilayer stacks seeks to infer layer materials, thicknesses, and ordering from a desired target spectrum. It is a long-standing challenge due to the large design space and non-unique solutions. We introduce \texttt{OptoLlama}, a masked diffusion language model for inverse thin-film design from optical spectra. Representing multilayer stacks as sequences of material-thickness tokens, \texttt{OptoLlama} conditions generation on reflectance, absorptance, and transmittance spectra and learns a probabilistic mapping from optical response to structure. Evaluated on a representative test set of 3,000 targets, \texttt{OptoLlama} reduces the mean absolute spectral error by 2.9-fold relative to a nearest-neighbor template baseline and by 3.45-fold relative to the state-of-the-art data-driven baseline, called \texttt{OptoGPT}. Case studies on designed and expert-defined targets show that the model reproduces characteristic spectral features and recovers physically meaningful stack motifs, including distributed Bragg reflectors. These results establish diffusion-based sequence modeling as a powerful framework for inverse photonic design.

physics.optics↗

arXiv2026

Differentiable Power-Flow Optimization

With the rise of renewable energy sources and their high variability in generation, the management of power grids becomes increasingly complex and computationally demanding. Conventional AC-power-flow simulations, which use the Newton-Raphson (NR) method, suffer from poor scalability, making them impractical for emerging use cases such as joint transmission-distribution modeling and global grid analysis. At the same time, purely data-driven surrogate models lack physical guarantees and may violate fundamental constraints. In this work, we propose Differentiable Power-Flow (DPF), a reformulation of the AC power-flow problem as a differentiable simulation. DPF enables end-to-end gradient propagation from the physical power mismatches to the underlying simulation parameters, thereby allowing these parameters to be identified efficiently using gradient-based optimization. We demonstrate that DPF provides a scalable alternative to NR by leveraging GPU acceleration, sparse tensor representations, and batching capabilities available in modern machine-learning frameworks such as PyTorch. DPF is especially suited as a tool for time-series analyses due to its efficient reuse of previous solutions, for N-1 contingency-analyses due to its ability to process cases in batches, and as a screening tool by leveraging its speed and early stopping capability. The code is available in the authors' code repository.

cs.AI↗

arXiv2026

Bayesian Lottery Ticket Hypothesis

Bayesian neural networks (BNNs) are a useful tool for uncertainty quantification, but require substantially more computational resources than conventional neural networks. For non-Bayesian networks, the Lottery Ticket Hypothesis (LTH) posits the existence of sparse subnetworks that can train to the same or even surpassing accuracy as the original dense network. Such sparse networks can lower the demand for computational resources at inference, and during training. The existence of the LTH and corresponding sparse subnetworks in BNNs could motivate the development of sparse training algorithms and provide valuable insights into the underlying training process. Towards this end, we translate the LTH experiments to a Bayesian setting using common computer vision models. We investigate the defining characteristics of Bayesian lottery tickets, and extend our study towards a transplantation method connecting BNNs with deterministic Lottery Tickets. We generally find that the LTH holds in BNNs, and winning tickets of matching and surpassing accuracy are present independent of model size, with degradation at very high sparsities. However, the pruning strategy should rely primarily on magnitude, secondly on standard deviation. Furthermore, our results demonstrate that models rely on mask structure and weight initialization to varying degrees.

cs.LG↗

arXiv2026

Feed-Forward Optimization With Delayed Feedback for Neural Network Training

Backpropagation has long been criticized for being biologically implausible due to its reliance on concepts that are not viable in natural learning processes. Two core issues are the weight transport and update locking problems caused by the forward-backward dependencies, which limit biological plausibility, computational efficiency, and parallelization. Although several alternatives have been proposed to increase biological plausibility, they often come at the cost of reduced predictive performance. This paper proposes an alternative approach to training feed-forward neural networks addressing these issues by using approximate gradient information. We introduce Feed-Forward with delayed Feedback (F$^3$), which approximates gradients using fixed random feedback paths and delayed error information from the previous epoch to balance biological plausibility with predictive performance. We evaluate F$^3$ across multiple tasks and architectures, including both fully-connected and Transformer networks. Our results demonstrate that, compared to similarly plausible approaches, F$^3$ significantly improves predictive performance, narrowing the gap to backpropagation by up to 56% for classification and 96% for regression. This work is a step towards more biologically plausible learning algorithms while opening up new avenues for energy-efficient and parallelizable neural network training.

cs.LG↗

arXiv2026

Beyond Backpropagation: Optimization with Multi-Tangent Forward Gradients

The gradients used to train neural networks are typically computed using backpropagation. While an efficient way to obtain exact gradients, backpropagation is computationally expensive, hinders parallelization, and is biologically implausible. Forward gradients are an approach to approximate the gradients from directional derivatives along random tangents computed by forward-mode automatic differentiation. So far, research has focused on using a single tangent per step. This paper provides an in-depth analysis of multi-tangent forward gradients and introduces an improved approach to combining the forward gradients from multiple tangents based on orthogonal projections. We demonstrate that increasing the number of tangents improves both approximation quality and optimization performance across various tasks.

cs.LG↗

arXiv2025

Exploring Federated Learning for Thermal Urban Feature Segmentation -- A Comparison of Centralized and Decentralized Approaches

Federated Learning (FL) is an approach for training a shared Machine Learning (ML) model with distributed training data and multiple participants. FL allows bypassing limitations of the traditional Centralized Machine Learning CL if data cannot be shared or stored centrally due to privacy or technical restrictions -- the participants train the model locally with their training data and do not need to share it among the other participants. This paper investigates the practical implementation and effectiveness of FL in a real-world scenario, specifically focusing on unmanned aerial vehicle (UAV)-based thermal images for common thermal feature detection in urban environments. The distributed nature of the data arises naturally and makes it suitable for FL applications, as images captured in two German cities are available. This application presents unique challenges due to non-identical distribution and feature characteristics of data captured at both locations. The study makes several key contributions by evaluating FL algorithms in real deployment scenarios rather than simulation. We compare several FL approaches with a centralized learning baseline across key performance metrics such as model accuracy, training time, communication overhead, and energy usage. This paper also explores various FL workflows, comparing client-controlled workflows and server-controlled workflows. The findings of this work serve as a valuable reference for understanding the practical application and limitations of the FL methods in segmentation tasks in UAV-based imaging.

cs.LG↗

arXiv2025

pyGinkgo: A Sparse Linear Algebra Operator Framework for Python

Sparse linear algebra is a cornerstone of many scientific computing and machine learning applications. Python has become a popular choice for these applications due to its simplicity and ease of use. Yet high performance sparse kernels in Python remain limited in functionality, especially on modern CPU and GPU architectures. We present pyGinkgo, a lightweight and Pythonic interface to the Ginkgo library, offering high-performance sparse linear algebra support with platform portability across CUDA, HIP, and OpenMP backends. pyGinkgo bridges the gap between high-performance C++ backends and Python usability by exposing Ginkgo's capabilities via Pybind11 and a NumPy and PyTorch compatible interface. We benchmark pyGinkgo's performance against state-of-the-art Python libraries including SciPy, CuPy, PyTorch, and TensorFlow. Results across hardware from different vendors demonstrate that pyGinkgo consistently outperforms existing Python tools in both sparse matrix vector (SpMV) product and iterative solver performance, while maintaining performance parity with native Ginkgo C++ code. Our work positions pyGinkgo as a compelling backend for sparse machine learning models and scientific workflows.

cs.MS↗

arXiv2025

Energy Consumption in Parallel Neural Network Training

The increasing demand for computational resources of training neural networks leads to a concerning growth in energy consumption. While parallelization has enabled upscaling model and dataset sizes and accelerated training, its impact on energy consumption is often overlooked. To close this research gap, we conducted scaling experiments for data-parallel training of two models, ResNet50 and FourCastNet, and evaluated the impact of parallelization parameters, i.e., GPU count, global batch size, and local batch size, on predictive performance, training time, and energy consumption. We show that energy consumption scales approximately linearly with the consumed resources, i.e., GPU hours; however, the respective scaling factor differs substantially between distinct model trainings and hardware, and is systematically influenced by the number of samples and gradient updates per GPU hour. Our results shed light on the complex interplay of scaling up neural network training and can inform future developments towards more sustainable AI research.

cs.LG↗

arXiv2025

Jigsaw: Training Multi-Billion-Parameter AI Weather Models with Optimized Model Parallelism

AI-based methods have revolutionized atmospheric forecasting, with recent successes in medium-range forecasting spurring the development of climate foundation models. Accurate modeling of complex atmospheric dynamics at high spatial resolutions and longer lead times requires large neural networks and gigabyte-sized data samples, making accelerator memory and I/O-bandwidth the bottlenecks for model training. We introduce WeatherMixer, a multi-layer-perceptron-based architecture whose workload scales linearly with input size, allowing the model to learn global weather phenomena at accuracies similar to numerical weather prediction. To cope with the computational demand, we propose Jigsaw, a novel model parallelization scheme that employs both domain and tensor parallelism, eliminating memory redundancy. Jigsaw exceeds state-of-the-art performance in strong scaling in compute-communication-limited systems and achieves superscalar weak scaling in I/O-bandwidth-limited systems. We scale training to 256 GPUs, reaching peak performances of 9 and 11 PFLOPs, 23% and 28% of theoretical peaks, achieving 68% and 72% scaling efficiency versus 51% without model parallelism.

cs.LG↗

arXiv2025

PETNet -- Coincident Particle Event Detection using Spiking Neural Networks

Spiking neural networks (SNN) hold the promise of being a more biologically plausible, low-energy alternative to conventional artificial neural networks. Their time-variant nature makes them particularly suitable for processing time-resolved, sparse binary data. In this paper, we investigate the potential of leveraging SNNs for the detection of photon coincidences in positron emission tomography (PET) data. PET is a medical imaging technique based on injecting a patient with a radioactive tracer and detecting the emitted photons. One central post-processing task for inferring an image of the tracer distribution is the filtering of invalid hits occurring due to e.g. absorption or scattering processes. Our approach, coined PETNet, interprets the detector hits as a binary-valued spike train and learns to identify photon coincidence pairs in a supervised manner. We introduce a dedicated multi-objective loss function and demonstrate the effects of explicitly modeling the detector geometry on simulation data for two use-cases. Our results show that PETNet can outperform the state-of-the-art classical algorithm with a maximal coincidence detection $F_1$ of 95.2%. At the same time, PETNet is able to predict photon coincidences up to 36 times faster than the classical approach, highlighting the great potential of SNNs in particle physics applications.

cs.LG↗

arXiv2025

Model Fusion via Neuron Transplantation

Ensemble learning is a widespread technique to improve the prediction performance of neural networks. However, it comes at the price of increased memory and inference time. In this work we propose a novel model fusion technique called \emph{Neuron Transplantation (NT)} in which we fuse an ensemble of models by transplanting important neurons from all ensemble members into the vacant space obtained by pruning insignificant neurons. An initial loss in performance post-transplantation can be quickly recovered via fine-tuning, consistently outperforming individual ensemble members of the same model capacity and architecture. Furthermore, NT enables all the ensemble members to be jointly pruned and jointly trained in a combined model. Comparing it to alignment-based averaging (like Optimal-Transport-fusion), it requires less fine-tuning than the corresponding OT-fused model, the fusion itself is faster and requires less memory, while the resulting model performance is comparable or better. The code is available under the following link: https://github.com/masterbaer/neuron-transplantation.

cs.LG↗

arXiv2024

A Comparative Study of Pruning Methods in Transformer-based Time Series Forecasting

The current landscape in time-series forecasting is dominated by Transformer-based models. Their high parameter count and corresponding demand in computational resources pose a challenge to real-world deployment, especially for commercial and scientific applications with low-power embedded devices. Pruning is an established approach to reduce neural network parameter count and save compute. However, the implications and benefits of pruning Transformer-based models for time series forecasting are largely unknown. To close this gap, we provide a comparative benchmark study by evaluating unstructured and structured pruning on various state-of-the-art multivariate time series models. We study the effects of these pruning strategies on model predictive performance and computational aspects like model size, operations, and inference time. Our results show that certain models can be pruned even up to high sparsity levels, outperforming their dense counterpart. However, fine-tuning pruned models is necessary. Furthermore, we demonstrate that even with corresponding hardware and software support, structured pruning is unable to provide significant time savings.

cs.LG↗

arXiv2024

AutoPQ: Automating Quantile estimation from Point forecasts in the context of sustainability

Optimizing smart grid operations relies on critical decision-making informed by uncertainty quantification, making probabilistic forecasting a vital tool. Designing such forecasting models involves three key challenges: accurate and unbiased uncertainty quantification, workload reduction for data scientists during the design process, and limitation of the environmental impact of model training. In order to address these challenges, we introduce AutoPQ, a novel method designed to automate and optimize probabilistic forecasting for smart grid applications. AutoPQ enhances forecast uncertainty quantification by generating quantile forecasts from an existing point forecast by using a conditional Invertible Neural Network (cINN). AutoPQ also automates the selection of the underlying point forecasting method and the optimization of hyperparameters, ensuring that the best model and configuration is chosen for each application. For flexible adaptation to various performance needs and available computing power, AutoPQ comes with a default and an advanced configuration, making it suitable for a wide range of smart grid applications. Additionally, AutoPQ provides transparency regarding the electricity consumption required for performance improvements. We show that AutoPQ outperforms state-of-the-art probabilistic forecasting methods while effectively limiting computational effort and hence environmental impact. Additionally and in the context of sustainability, we quantify the electricity consumption required for performance improvements.

cs.LG↗

arXiv2024

Harnessing Orthogonality to Train Low-Rank Neural Networks

This study explores the learning dynamics of neural networks by analyzing the singular value decomposition (SVD) of their weights throughout training. Our investigation reveals that an orthogonal basis within each multidimensional weight's SVD representation stabilizes during training. Building upon this, we introduce Orthogonality-Informed Adaptive Low-Rank (OIALR) training, a novel training method exploiting the intrinsic orthogonality of neural networks. OIALR seamlessly integrates into existing training workflows with minimal accuracy loss, as demonstrated by benchmarking on various datasets and well-established network architectures. With appropriate hyperparameter tuning, OIALR can surpass conventional training setups, including those of state-of-the-art models.

cs.LG↗

arXiv2024

AB-Training: A Communication-Efficient Approach for Distributed Low-Rank Learning

Communication bottlenecks severely hinder the scalability of distributed neural network training, particularly in high-performance computing (HPC) environments. We introduce AB-training, a novel data-parallel method that leverages low-rank representations and independent training groups to significantly reduce communication overhead. Our experiments demonstrate an average reduction in network traffic of approximately 70.31\% across various scaling scenarios, increasing the training potential of communication-constrained systems and accelerating convergence at scale. AB-training also exhibits a pronounced regularization effect at smaller scales, leading to improved generalization while maintaining or even reducing training time. We achieve a remarkable 44.14 : 1 compression ratio on VGG16 trained on CIFAR-10 with minimal accuracy loss, and outperform traditional data parallel training by 1.55\% on ResNet-50 trained on ImageNet-2012. While AB-training is promising, our findings also reveal that large batch effects persist even in low-rank regimes, underscoring the need for further research into optimized update mechanisms for massively distributed training.

cs.LG↗

arXiv2024

ReCycle: Fast and Efficient Long Time Series Forecasting with Residual Cyclic Transformers

Transformers have recently gained prominence in long time series forecasting by elevating accuracies in a variety of use cases. Regrettably, in the race for better predictive performance the overhead of model architectures has grown onerous, leading to models with computational demand infeasible for most practical applications. To bridge the gap between high method complexity and realistic computational resources, we introduce the Residual Cyclic Transformer, ReCycle. ReCycle utilizes primary cycle compression to address the computational complexity of the attention mechanism in long time series. By learning residuals from refined smoothing average techniques, ReCycle surpasses state-of-the-art accuracy in a variety of application use cases. The reliable and explainable fallback behavior ensured by simple, yet robust, smoothing average techniques additionally lowers the barrier for user acceptance. At the same time, our approach reduces the run time and energy consumption by more than an order of magnitude, making both training and inference feasible on low-performance, low-power and edge computing devices. Code is available at https://github.com/Helmholtz-AI-Energy/ReCycle

cs.LG↗