6 papers
Distributional Loss for Robust Classification
Kathleen Anderson, Thomas Martinetz
This paper proposes a novel loss concept for supervised classification tasks. Rather than enforcing a direct mapping from each input sample to a single assigned label, we define an…
Nearest-Neighbor Density Estimation for Dependency Suppression
Kathleen Anderson, Thomas Martinetz
The ability to remove unwanted dependencies from data is crucial in various domains, including fairness, robust learning, and privacy protection. In this work, we propose an encode…
The Resurrection of the ReLU
CoÅku Can Horuz, Geoffrey Kasenbacher, Saya Higuchi +7
Modeling sophisticated activation functions within deep learning architectures has evolved into a distinct research direction. Functions such as GELU, SELU, and SiLU offer smooth g…
Revealing Unintentional Information Leakage in Low-Dimensional Facial Portrait Representations
Kathleen Anderson, Thomas Martinetz
We evaluate the information that can unintentionally leak into the low dimensional output of a neural network, by reconstructing an input image from a 40- or 32-element feature vec…
Why LLMs Cannot Think and How to Fix It
Marius Jahrens, Thomas Martinetz
This paper elucidates that current state-of-the-art Large Language Models (LLMs) are fundamentally incapable of making decisions or developing "thoughts" within the feature space d…
Enhancing Generalization in Convolutional Neural Networks through Regularization with Edge and Line Features
Christoph Linse, Beatrice Brückner, Thomas Martinetz
This paper proposes a novel regularization approach to bias Convolutional Neural Networks (CNNs) toward utilizing edge and line features in their hidden layers. Rather than learnin…