Zobrazeno 1 - 10
of 4 329
pro vyhledávání: '"A. Flügel"'
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. Forwar
Externí odkaz:
http://arxiv.org/abs/2410.17764
Deep learning models are often unaware of the inherent constraints of the task they are applied to. However, many downstream tasks require logical consistency. For ontology classification tasks, such constraints include subsumption and disjointness r
Externí odkaz:
http://arxiv.org/abs/2405.02083
Autor:
Coquelin, Daniel, Flügel, Katherina, Weiel, Marie, Kiefer, Nicholas, Öz, Muhammed, Debus, Charlotte, Streit, Achim, Götz, Markus
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 representa
Externí odkaz:
http://arxiv.org/abs/2405.01067
Autor:
Coquelin, Daniel, Flügel, Katharina, Weiel, Marie, Kiefer, Nicholas, Debus, Charlotte, Streit, Achim, Götz, Markus
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 repres
Externí odkaz:
http://arxiv.org/abs/2401.08505
Publikováno v:
Current Directions in Biomedical Engineering, Vol 10, Iss 1, Pp 65-68 (2024)
A deformation of the hard palate can occur in spinal muscular atrophy and leads to problems with feeding and swallowing in early childhood. An objective analysis of the palatal changes is therefore desirable for early treatment initiation. In this st
Externí odkaz:
https://doaj.org/article/6db6d0e4ac644681b4d515a97ff82d9f
Autor:
Flügel, Katharina, Coquelin, Daniel, Weiel, Marie, Debus, Charlotte, Streit, Achim, Götz, Markus
Backpropagation has long been criticized for being biologically implausible, relying on concepts that are not viable in natural learning processes. This paper proposes an alternative approach to solve two core issues, i.e., weight transport and updat
Externí odkaz:
http://arxiv.org/abs/2304.13372
Autor:
Matthias Münchhalfen, Richard Görg, Michael Haberl, Jens Löber, Jakob Willenbrink, Laura Schwarzt, Charlotte Höltermann, Christian Ickes, Leonard Hammermann, Jan Kus, Björn Chapuy, Andrea Ballabio, Sybille D. Reichardt, Alexander Flügel, Niklas Engels, Jürgen Wienands
Publikováno v:
Nature Communications, Vol 15, Iss 1, Pp 1-21 (2024)
Abstract Ligation of the B cell antigen receptor (BCR) initiates humoral immunity. However, BCR signaling without appropriate co-stimulation commits B cells to death rather than to differentiation into immune effector cells. How BCR activation deplet
Externí odkaz:
https://doaj.org/article/d3d3f15046064c9eadb2e5e5e120d8db
In ontology development, there is a gap between domain ontologies which mostly use the web ontology language, OWL, and foundational ontologies written in first-order logic, FOL. To bridge this gap, we present Gavel, a tool that supports the developme
Externí odkaz:
http://arxiv.org/abs/2210.03497
Autor:
Nelson, Jocienne N., Rice, Anthony D., Brooks, Chase, Leahy, Ian A., Teeter, Glenn, Van Schilfgaarde, Mark, Lany, Stephan, Fluegel, Brian, Lee, Minhyea, Alberi, Kirstin
Publikováno v:
Phys. Rev. B 107, L220206 2023
The extent to which disorder influences the properties of topological semimetals remains an open question and is relevant to both the understanding of topological states and the use of topological materials in practical applications. Here, we achieve
Externí odkaz:
http://arxiv.org/abs/2206.10023
CUBES is the Cassegrain U-Band Efficient Spectrograph, a high-efficiency instrument operating in the UV spectral range between 300nm and 400nm with a resolution not less than 20000. CUBES is to be installed at a Cassegrain focus of the Very Large Tel
Externí odkaz:
http://arxiv.org/abs/2203.15637