no code implementations • 17 Mar 2023 • Andreas Lohrer, Darpan Malik, Claudius Zelenka, Peer Kröger
Hence, this paper introduces GADformer, a BERT-based model for attention-driven GAD on trajectories in unsupervised and semi-supervised settings.
no code implementations • 23 Feb 2023 • Marco Landt-Hayen, Willi Rath, Martin Claus, Peer Kröger
LRP is often used in the context of image classification.
no code implementations • 23 Nov 2022 • Andreas Lohrer, Daniyal Kazempour, Maximilian Hünemörder, Peer Kröger
Unsupervised learning methods are well established in the area of anomaly detection and achieve state of the art performances on outlier data sets.
no code implementations • 18 Oct 2022 • Marco Landt-Hayen, Peer Kröger, Martin Claus, Willi Rath
We also show how ESNs can be used not only for time series prediction but also for image classification: Our ESN model serves as a detector for El Nino Southern Oscillation (ENSO) from sea surface temperature anomalies.
1 code implementation • 1 Jul 2022 • Moritz Herrmann, Daniyal Kazempour, Fabian Scheipl, Peer Kröger
We discuss topological aspects of cluster analysis and show that inferring the topological structure of a dataset before clustering it can considerably enhance cluster detection: theoretical arguments and empirical evidence show that clustering embedding vectors, representing the structure of a data manifold instead of the observed feature vectors themselves, is highly beneficial.
no code implementations • 8 Sep 2021 • Martin Ringsquandl, Evgeny Kharlamov, Daria Stepanova, Steffen Lamparter, Raffaello Lepratti, Ian Horrocks, Peer Kröger
Smooth operation of such factories requires that the machines and engineering personnel that conduct their monitoring and diagnostics share a detailed common industrial knowledge about the factory, e. g., in the form of knowledge graphs.
no code implementations • 18 Dec 2014 • Johannes Niedermayer, Peer Kröger
In this paper we address an approach complementary to indexing in order to improve the runtimes of retrieval by querying only the most promising keypoint descriptors, as this affects matching runtimes linearly and can therefore lead to increased efficiency.