Prof. Dr. Marina Höhne
Head of Department
Department: Data Science in Bioeconomy
Telefon: +49 331 5699-902
E-Mail: MHoehne@atb-potsdam.de
Online:
Research program
Committees and boards
- Climate Change and AI in Brandenburg center
- ELLIS Society, the European Laboratory for Learning and Intelligent Systems
- Berlin AI Competence Center BIFOLD, the Berlin Institute for the Foundations of Learning and Data
Projects
- SPIN-FERT – Innovative practices, tools and products to boost soil fertility and peat substitution in horticultural crops The main objective of SPIN-FERT is to integrate optimised and validated innovations in soil management practices and improve pea…
- DCropS4OneHealth-2 – Diversifying cropping systems for the One Health of soils, plants and humans Die Diversifizierung von Pflanzenbausystemen in Verbindung mit der Senkung des Einsatzes von Agrochemikalien und einer Erhöhung der Landschaftskomplexit…
- Joint Lab KI & DS – Joint Lab Künstliche Intelligenz & Data Science Im Rahmen des Joint Lab bilden die Universität Osnabrück und das ATB gemeinsam Doktorandinnen und Doktoranden an der Schnittstelle von Agrarwissenschaft und Künstlicher Intelligenz a…
- Explaining 4.0 – Künstliche Intelligenz - Transparenz und Effizienz Das Ziel des Projektes Explaining 4.0 ist die Entwicklung von Methoden, die einen signifikanten Beitrag zu einem ganzheitlichen -globalen- Verständnis von KI-Modellen leisten. Dabei …
- DCropS4OneHealth-1 – Diversifying cropping systems for the One Health of soils, plants and humans; BiodivGesundheit: Diversifizierung von Pflanzenbausystemen für die gemeinsame Gesundheit von Böden, Pflanzen und Menschen Die Diversifizierung von Pfl…
- XAI-Mobil – Towards Reliable Artificial Intelligence for Explainable, Interactive and Self-evolving Systems Projekt im Rahmen des Mobilitätsförderprogramms des Chinesisch-Deutschen Zentrum für Wissenschaftsförderung mit Forschern der Sun Yat-sen Univ…
Publications
- Bommer, P.; Kretschmer, M.; Hedström, A.; Bareeva, D.; Höhne, M. (2024): Finding the right XAI Method - A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate Science. Artificial Intelligence for the Earth Systems (AIES). : p. 1-55. Online: https://doi.org/10.1175/AIES-D-23-0074.1
- Bareeva, D.; Höhne, M.; Warnecke, A.; Pirch, L.; Müller, K.; Rieck, K.; Bykov, K. (2024): Manipulating Feature Visualizations with Gradient Slingshots. arXiv. : p. 1-19. Online: https://doi.org/10.48550/arXiv.2401.06122
- Gautam, S.; Boubekki, A.; Höhne, M.; Kampffmeyer, M. (2023): Prototypical Self-Explainable Models Without Re-training. arXiv. : p. 1-25. Online: https://arxiv.org/abs/2312.07822
- Hedström, A.; Weber, L.; Lapuschkin, S.; Höhne, M. (2023): Sanity Checks Revisited: An Exploration to Repair the Model Parameter Randomisation Test. arXiv. : p. 1-19. Online: https://arxiv.org/abs/2401.06465
- Bykov, K.; Kopf, L.; Nakajima, S.; Kloft, M.; Höhne, M. (2023): Labeling Neural Representations with Inverse Recognition. arXiv. : p. 1-24. Online: https://doi.org/10.48550/arXiv.2311.13594
- Grinwald, D.; Bykov, K.; Nakajima, S.; Höhne, M. (2023): Visualizing the Diversity of Representations Learned by Bayesian Neural Networks. Transactions on Machine Learning Research. (11): p. 1-25. Online: https://openreview.net/pdf?id=ZSxvyWrX6k
- Hanfeld, P.; Wahba, K.; Höhne, M.; Bussmann, M.; Hönig, W. (2023): Kidnapping Deep Learning-based Multirotors using Optimized Flying Adversarial Patches. arXiv. : p. 1-7. Online: https://doi.org/10.48550/arXiv.2308.00344
- Hanfeld, P.; Höhne, M.; Bussmann, M.; Hönig, W. (2023): Flying Adversarial Patches: Manipulating the Behavior of Deep Learning-based Autonomous Multirotors. arXiv. : p. 1-6. Online: https://doi.org/10.48550/arXiv.2305.12859
- Hedström, A.; Bommer, P.; Wickstrøm, K.; Samek, W.; Lapuschkin, S.; Höhne, M. (2023): The Meta-Evaluation Problem in Explainable AI: Identifying Reliable Estimators with MetaQuantus. Transactions on Machine Learning Research. (06): p. 1-35. Online: https://openreview.net/forum?id=j3FK00HyfU
- Bykov, K.; Deb, M.; Grinwald, D.; Müller, K.; Höhne, M. (2023): DORA: Exploring Outlier Representations in Deep Neural Networks. Transactions on Machine Learning Research. (06): p. 1-43. Online: https://doi.org/10.48550/arXiv.2206.04530
Weitere Veröffentlichungen
- Gautam, S., Höhne, M. M.-C., Hansen, S., Jenssen, R., & Kampffmeyer, M. (2022). Demonstrating The Risk of Imbalanced Datasets in Chest X-ray Image-based Diagnostics by Prototypical Relevance Propagation. In 2022 IEEE 19th International Symposium on Biomedical Imaging (ISBI) (pp. 1-5). IEEE
- Hedström, A., Weber, L., Bareeva, D., Motzkus, F., Samek, W., Lapuschkin, S., and Höhne, M. M.-C. (2022). Quantus: an explainable AI toolkit for responsible evaluation of neural network explanations. arXiv preprint arXiv:2202.06861
- Mieth, B., Rozier, A., Rodriguez, J. A., Höhne, M.M.-C., Görnitz, N., & Müller, K. R. (2021): DeepCOMBI: explainable artificial intelligence for the analysis and discovery in genome-wide association studies. NAR Genomics and Bioinformatics, 3(3), lqab065
- Bykov, K., Deb, M., Grinwald, D., Müller, K.R. and Höhne, M.M.-C., 2022. DORA: Exploring outlier representations in Deep Neural Networks. arXiv preprint arXiv: 2206.04530
- Bykov, K., Hedström, A., Nakajima, S., and Höhne, M.M.-C. (2021): NoiseGrad: enhancing explanations by introducing stochasticity to model weights. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 36, No. 6, pp. 6132-6140)
- Bykov, K., Höhne, M.M.-C., Creosteanu, A., Müller, K.-R., Klauschen, F., Nakajima, S., & Kloft, M. (2021). Explaining bayesian neural networks. arXiv preprint arXiv:2108.10346
- Mieth, B., Hockley, J. R., Görnitz, N., Vidovic, M.M.-C., Müller, K. R., Gutteridge, A., and Ziemek, D. (2019): Using transfer learning from prior reference knowledge to improve the clustering of single-cell RNA-Seq data. Scientific reports, 9(1), 1-14
- Vidovic, M.M.-C., Kloft M., Müller K.-R., and Görnitz N., 2017. ML2motif – Reliable extraction of discriminative sequence motifs from learning machines. PloS one 12.3, e0174392
- Vidovic M.M.-C., Görnitz N., Müller K.-R., and Kloft M., 2016. Feature importance measure for non-linear learning algorithms. arXiv preprint arXiv:1611.07567
- Vidovic, M.M.-C., Hwang H.J., Amsüss S., Hahne J.M., Farina D., and Müller K.-R. (2016): Improving the robustness of myoelectric pattern recognition for upper limb prostheses by covariate shift adaptation. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 24.9, 961-970
- Vidovic, M.M.-C., Görnitz N., Müller K.-R., Rätsch G., and Kloft M. (2015): SVM2Motif Reconstructing Overlapping DNA Sequence Motifs by Mimicking an SVM Predictor. PloS one 10.12, e0144782