no code implementations • 27 Aug 2023 • Bettina Finzel, Simon P. Kuhn, David E. Tafler, Ute Schmid
In this paper, we present an approach for generating contrastive explanations to explain facial expressions of pain and disgust shown in video sequences.
no code implementations • 4 Jul 2023 • Jonas-Dario Troles, Richard Nieding, Sonia Simons, Ute Schmid
Furthermore, Bamberg already has a georeferenced tree cadastre of around 15, 000 solitary trees in the city area, which is also used to generate helpful information.
no code implementations • 30 May 2023 • Durgesh Nandini, Ute Schmid
This has been achieved by first predicting the classification of a text and then providing a posthoc, model agnostic and surrogate interpretability approach for explainability and to prevent model bias.
no code implementations • 31 Oct 2022 • Ines Rieger, Jaspar Pahl, Bettina Finzel, Ute Schmid
Neural networks are widely adopted, yet the integration of domain knowledge is still underutilized.
1 code implementation • 20 May 2022 • Lun Ai, Johannes Langer, Stephen H. Muggleton, Ute Schmid
We propose a framework for the effects of sequential teaching on comprehension based on an existing definition of comprehensibility and provide evidence for support from data collected in human trials.
no code implementations • 6 Apr 2022 • Emanuel Slany, Yannik Ott, Stephan Scheele, Jan Paulus, Ute Schmid
This accuracy is approximately equal to state-of-the-art Deep Learning optimization procedures.
no code implementations • 4 Apr 2022 • Christoph Wehner, Francis Powlesland, Bashar Altakrouri, Ute Schmid
The core implementation includes consuming live data from a digital twin on a German highway, live predictions and explanations of lane changes by extending LRP to layer normalized LSTMs, and an interface for communicating and explaining the predictions to a human user.
no code implementations • 17 Mar 2022 • Dennis Müller, Michael März, Stephan Scheele, Ute Schmid
Machine learning based image classification algorithms, such as deep neural network approaches, will be increasingly employed in critical settings such as quality control in industry, where transparency and comprehensibility of decisions are crucial.
no code implementations • 3 Jan 2022 • Gesina Schwalbe, Christian Wirth, Ute Schmid
In this work, we present a simple, yet effective, approach to verify that a CNN complies with symbolic predicate logic rules which relate visual concepts.
no code implementations • 7 Oct 2021 • Bettina Finzel, David E. Tafler, Stephan Scheele, Ute Schmid
We present a process-based approach that combines multi-level and multi-modal explanations.
Explainable Artificial Intelligence (XAI) Inductive logic programming +1
no code implementations • WMT (EMNLP) 2021 • Jonas-Dario Troles, Ute Schmid
In this set measurement of gender bias is solely based on the translation of occupations.
no code implementations • 15 Jun 2021 • Johannes Rabold, Michael Siebers, Ute Schmid
Such near misses have been proposed by Winston (1970) as efficient guidance for learning in relational domains.
1 code implementation • 16 May 2021 • Johannes Rabold, Gesina Schwalbe, Ute Schmid
We show that our explanation is faithful to the original black-box model.
no code implementations • 9 Sep 2020 • Lun Ai, Stephen H. Muggleton, Céline Hocquette, Mark Gromowski, Ute Schmid
USML is demonstrated by a measurable increase in human performance of a task following provision to the human of a symbolic machine learned theory for task performance.
no code implementations • 14 Feb 2020 • Ines Rieger, Rene Kollmann, Bettina Finzel, Dominik Seuss, Ute Schmid
Finally, we quantify these visual explanations based on a bounding-box method with respect to facial regions.
Facial Expression Recognition Facial Expression Recognition (FER)
1 code implementation • 17 Oct 2019 • Ludwig Schallner, Johannes Rabold, Oliver Scholz, Ute Schmid
Quick-Shift resulted in the least and Compact-Watershed in the highest correspondence with the reference relevance areas.
no code implementations • 4 Oct 2019 • Johannes Rabold, Hannah Deininger, Michael Siebers, Ute Schmid
First, we show that our approach is capable of identifying a single relation as important explanatory construct.
no code implementations • 19 Dec 2014 • Mark Wernsdorfer, Ute Schmid
Explicit models of the environment can be learned to augment such a value function.
Hierarchical Reinforcement Learning reinforcement-learning +2