5 papers
Geometric and Information Compression of Representations in Deep Learning
Linara Adilova, Henning Petzka, Asja Fischer +1
Deep neural networks transform input data into latent representations that support a wide range of downstream tasks. These representations can be characterized along information-th…
Universal Outlier Hypothesis Testing via Mean- and Median-Based Tests
Bernhard C. Geiger, Tobias Koch, Josipa Mihaljević +1
Universal outlier hypothesis testing refers to a hypothesis testing problem where one observes a large number of length- sequences -- the majority of which are distributed accor…
On the Role of Priors in Bayesian Causal Learning
Bernhard C. Geiger, Roman Kern
In this work, we investigate causal learning of independent causal mechanisms from a Bayesian perspective. Confirming previous claims from the literature, we show in a didactically…
Constraining Anomaly Detection with Anomaly-Free Regions
Maximilian Toller, Hussain Hussain, Roman Kern +1
We propose the novel concept of anomaly-free regions (AFR) to improve anomaly detection. An AFR is a region in the data space for which it is known that there are no anomalies insi…
Activation Bottleneck: Sigmoidal Neural Networks Cannot Forecast a Straight Line
Maximilian Toller, Hussain Hussain, Bernhard C Geiger
A neural network has an activation bottleneck if one of its hidden layers has a bounded image. We show that networks with an activation bottleneck cannot forecast unbounded sequenc…