Michael Kirchhof
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Michael Kirchhof has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
Bálint Mucsányi
1 shared publication- Christopher Reining
Technische Universität Dortmund
1 shared publication Enkelejda Kasneci
1 shared publication- Enkelejda Kasneci
Technical University of Munich, Munich, Germany and Munich Center for Machine Learning (MCML), Munich, Germany
1 shared publication - Karsten Roth
University of Tübingen
1 shared publication - Lena Schmid
Department of Statistics TU Dortmund University
1 shared publication
Latest Publications
- Open Access
URL: A Representation Learning Benchmark for Transferable Uncertainty Estimates
Representation learning has significantly driven the field to develop pretrained models that can act as a valuable starting point when transferring to new datasets. With the rising demand for reliable machine learning and uncertainty quantification, there is a need for pretrained models that not only provide embeddings but also transferable uncertainty estimates. To guide the development of such models, we propose the Uncertainty-aware Representation Learning (URL) benchmark. Besides the transferability of the representations, it also measures the zero-shot transferability of the uncertainty estimate using a novel metric. We apply URL to evaluate eleven uncertainty quantifiers that are pretrained on ImageNet and transferred to eight downstream datasets. We find that approaches that focus on the uncertainty of the representation itself or estimate the prediction risk directly outperform those that are based on the probabilities of upstream classes. Yet, achieving transferable uncertainty quantification remains an open challenge. Our findings indicate that it is not necessarily in conflict with traditional representation learning goals. Code is provided under https://github.com/mkirchhof/url .
Open AccessURL: A Representation Learning Benchmark for Transferable Uncertainty Estimates
Representation learning has significantly driven the field to develop pretrained models that can act as a valuable starting point when transferring to new datasets. With the rising demand for reliable machine learning and uncertainty quantification, there is a need for pretrained models that not only provide embeddings but also transferable uncertainty estimates. To guide the development of such models, we propose the Uncertainty-aware Representation Learning (URL) benchmark. Besides the transferability of the representations, it also measures the zero-shot transferability of the uncertainty estimate using a novel metric. We apply URL to evaluate eleven uncertainty quantifiers that are pretrained on ImageNet and transferred to eight downstream datasets. We find that approaches that focus on the uncertainty of the representation itself or estimate the prediction risk directly outperform those that are based on the probabilities of upstream classes. Yet, achieving transferable uncertainty quantification remains an open challenge. Our findings indicate that it is not necessarily in conflict with traditional representation learning goals. Code is provided under https://github.com/mkirchhof/url .
- 1 November 2022
- Published by Springer Nature
- p. 435-454
A Non-isotropic Probabilistic Take on Proxy-based Deep Metric Learning
- 1 November 2022
- Published by Springer Nature
- p. 435-454
A Non-isotropic Probabilistic Take on Proxy-based Deep Metric Learning
- 22 September 2021
- Published by Springer Nature
- p. 163-177
Chances of Interpretable Transfer Learning for Human Activity Recognition in Warehousing
- 22 September 2021
- Published by Springer Nature
- p. 163-177
Chances of Interpretable Transfer Learning for Human Activity Recognition in Warehousing