arXiv subjects

Tobias Fuchs

Publications and source records attributed to Tobias Fuchs.

8 recordsLinked to original sources

arXiv2026

QDSB: Quantized Diffusion Schr\"odinger Bridges

Learning generative models in settings where the source and target distributions are only specified through unpaired samples is gaining in importance. Here, one frequently-used model are Schr\"odinger bridges (SB), which represent the most likely evolution between both endpoint distributions. To accelerate training, simulation-free SBs avoid the path simulation of the original SB models. However, learning simulation-free SBs requires paired data; a coupling of the source and target samples is obtained as the solution of the entropic optimal transport (OT) problem. As obtaining the optimal global coupling is infeasible in many practical cases, the entropic OT problem is iteratively solved on minibatches instead. Still, the repeated cost remains substantial and the locality can distort the global transport geometry. We propose quantized diffusion Schr\"odinger bridges (QDSB), which compute the endpoint coupling on anchor-quantized endpoint distributions and lift the resulting plan back to original data points through cell-wise sampling. We show that the regularized optimal coupling is stable w.r.t. anchor quantization, with an error controlled by the quality of the anchor approximation. In real-world experiments, QDSB matches the sample quality of existing baselines, requiring substantially less time. Code and data are available at github.com/mathefuchs/qdsb.

cs.LG↗

arXiv2025

Amortized Variational Inference for Partial-Label Learning: A Probabilistic Approach to Label Disambiguation

Real-world data is frequently noisy and ambiguous. In crowdsourcing, for example, human annotators may assign conflicting class labels to the same instances. Partial-label learning (PLL) addresses this challenge by training classifiers when each instance is associated with a set of candidate labels, only one of which is correct. While early PLL methods approximate the true label posterior, they are often computationally intensive. Recent deep learning approaches improve scalability but rely on surrogate losses and heuristic label refinement. We introduce a novel probabilistic framework that directly approximates the posterior distribution over true labels using amortized variational inference. Our method employs neural networks to predict variational parameters from input data, enabling efficient inference. This approach combines the expressiveness of deep learning with the rigor of probabilistic modeling, while remaining architecture-agnostic. Theoretical analysis and extensive experiments on synthetic and real-world datasets demonstrate that our method achieves state-of-the-art performance in both accuracy and efficiency.

cs.LG↗

arXiv2025

Robust Partial-Label Learning by Leveraging Class Activation Values

Real-world training data is often noisy; for example, human annotators assign conflicting class labels to the same instances. Partial-label learning (PLL) is a weakly supervised learning paradigm that allows training classifiers in this context without manual data cleaning. While state-of-the-art methods have good predictive performance, their predictions are sensitive to high noise levels, out-of-distribution data, and adversarial perturbations. We propose a novel PLL method based on subjective logic, which explicitly represents uncertainty by leveraging the magnitudes of the underlying neural network's class activation values. Thereby, we effectively incorporate prior knowledge about the class labels by using a novel label weight re-distribution strategy that we prove to be optimal. We empirically show that our method yields more robust predictions in terms of predictive performance under high PLL noise levels, handling out-of-distribution examples, and handling adversarial perturbations on the test instances.

cs.LG↗

arXiv2025

Partial-Label Learning with Conformal Candidate Cleaning

Real-world data is often ambiguous; for example, human annotation produces instances with multiple conflicting class labels. Partial-label learning (PLL) aims at training a classifier in this challenging setting, where each instance is associated with a set of candidate labels and one correct, but unknown, class label. A multitude of algorithms targeting this setting exists and, to enhance their prediction quality, several extensions that are applicable across a wide range of PLL methods have been introduced. While many of these extensions rely on heuristics, this article proposes a novel enhancing method that incrementally prunes candidate sets using conformal prediction. To work around the missing labeled validation set, which is typically required for conformal prediction, we propose a strategy that alternates between training a PLL classifier to label the validation set, leveraging these predicted class labels for calibration, and pruning candidate labels that are not part of the resulting conformal sets. In this sense, our method alternates between empirical risk minimization and candidate set pruning. We establish that our pruning method preserves the conformal validity with respect to the unknown ground truth. Our extensive experiments on artificial and real-world data show that the proposed approach significantly improves the test set accuracies of several state-of-the-art PLL classifiers.

cs.LG↗

arXiv2024

Milling using two mechatronically coupled robots

Industrial robots are commonly used in various industries due to their flexibility. However, their adoption for machining tasks is minimal because of the low dynamic stiffness characteristic of serial kinematic chains. To overcome this problem, we propose coupling two industrial robots at the flanges to form a parallel kinematic machining system. Although parallel kinematic chains are inherently stiffer, one possible disadvantage of the proposed system is that it is heavily overactuated. We perform a modal analysis to show that this may be an advantage, as the redundant degrees of freedom can be used to shift the natural frequencies by applying tension to the coupling module. To demonstrate the validity of our approach, we perform a milling experiment using our coupled system. An external measurement system is used to show that tensioning the coupling module causes a deformation of the system. We further show that this deformation is static over the tool path and can be compensated for.

cs.RO↗

arXiv2024

Partial-Label Learning with a Reject Option

In real-world applications, one often encounters ambiguously labeled data, where different annotators assign conflicting class labels. Partial-label learning allows training classifiers in this weakly supervised setting, where state-of-the-art methods already show good predictive performance. However, even the best algorithms give incorrect predictions, which can have severe consequences when they impact actions or decisions. We propose a novel risk-consistent nearest-neighbor-based partial-label learning algorithm with a reject option, that is, the algorithm can reject unsure predictions. Extensive experiments on artificial and real-world datasets show that our method provides the best trade-off between the number and accuracy of non-rejected predictions when compared to our competitors, which use confidence thresholds for rejecting unsure predictions. When evaluated without the reject option, our nearest-neighbor-based approach also achieves competitive prediction performance.

cs.LG↗

arXiv2016

DASH: A C++ PGAS Library for Distributed Data Structures and Parallel Algorithms

We present DASH, a C++ template library that offers distributed data structures and parallel algorithms and implements a compiler-free PGAS (partitioned global address space) approach. DASH offers many productivity and performance features such as global-view data structures, efficient support for the owner-computes model, flexible multidimensional data distribution schemes and inter-operability with STL (standard template library) algorithms. DASH also features a flexible representation of the parallel target machine and allows the exploitation of several hierarchically organized levels of locality through a concept of Teams. We evaluate DASH on a number of benchmark applications and we port a scientific proxy application using the MPI two-sided model to DASH. We find that DASH offers excellent productivity and performance and demonstrate scalability up to 9800 cores.

cs.DC↗

arXiv2016

Effective use of the PGAS Paradigm: Driving Transformations and Self-Adaptive Behavior in DASH-Applications

DASH is a library of distributed data structures and algorithms designed for running the applications on modern HPC architectures, composed of hierarchical network interconnections and stratified memory. DASH implements a PGAS (partitioned global address space) model in the form of C++ templates, built on top of DART -- a run-time system with an abstracted tier above existing one-sided communication libraries. In order to facilitate the application development process for exploiting the hierarchical organization of HPC machines, DART allows to reorder the placement of the computational units. In this paper we present an automatic, hierarchical units mapping technique (using a similar approach to the Hilbert curve transformation) to reorder the placement of DART units on the Cray XC40 machine Hazel Hen at HLRS. To evaluate the performance of new units mapping which takes into the account the topology of allocated compute nodes, we perform latency benchmark for a 3D stencil code. The technique of units mapping is generic and can be be adopted in other DART communication substrates and on other hardware platforms. Furthermore, high--level features of DASH are presented, enabling more complex automatic transformations and optimizations in the future.

cs.DC↗