Zobrazeno 1 - 10
of 26
pro vyhledávání: '"Goschenhofer Jann"'
Autor:
Dexl Jakob, Benz Michaela, Kuritcyn Petr, Wittenberg Thomas, Bruns Volker, Geppert Carol, Hartmann Arndt, Bischl Bernd, Goschenhofer Jann
Publikováno v:
Current Directions in Biomedical Engineering, Vol 8, Iss 2, Pp 344-347 (2022)
We explore the task of tissue classification for colon cancer histology in a low label regime comparing a semi-supervised and a supervised learning strategy in a series of experiments. Further, we investigate the model robustness w.r.t. distribution
Externí odkaz:
https://doaj.org/article/3a436ac07def4f1294c9db3081cb1688
Semi-supervised learning by self-training heavily relies on pseudo-label selection (PLS). The selection often depends on the initial model fit on labeled data. Early overfitting might thus be propagated to the final model by selecting instances with
Externí odkaz:
http://arxiv.org/abs/2302.08883
Autor:
Akkus, Cem, Chu, Luyang, Djakovic, Vladana, Jauch-Walser, Steffen, Koch, Philipp, Loss, Giacomo, Marquardt, Christopher, Moldovan, Marco, Sauter, Nadja, Schneider, Maximilian, Schulte, Rickmer, Urbanczyk, Karol, Goschenhofer, Jann, Heumann, Christian, Hvingelby, Rasmus, Schalk, Daniel, Aßenmacher, Matthias
This book is the result of a seminar in which we reviewed multimodal approaches and attempted to create a solid overview of the field, starting with the current state-of-the-art approaches in the two subfields of Deep Learning individually. Further,
Externí odkaz:
http://arxiv.org/abs/2301.04856
Positive-unlabeled learning (PUL) aims at learning a binary classifier from only positive and unlabeled training data. Even though real-world applications often involve imbalanced datasets where the majority of examples belong to one class, most cont
Externí odkaz:
http://arxiv.org/abs/2201.13192
Autor:
Goschenhofer, Jann, Hvingelby, Rasmus, Rügamer, David, Thomas, Janek, Wagner, Moritz, Bischl, Bernd
Publikováno v:
2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA)
While Semi-supervised learning has gained much attention in computer vision on image data, yet limited research exists on its applicability in the time series domain. In this work, we investigate the transferability of state-of-the-art deep semi-supe
Externí odkaz:
http://arxiv.org/abs/2102.03622
Parkinson's disease (PD) is the second most common neurodegenerative disease worldwide and affects around 1% of the (60+ years old) elderly population in industrial nations. More than 80% of PD patients suffer from motor symptoms, which could be well
Externí odkaz:
http://arxiv.org/abs/1911.06913
Autor:
Goschenhofer, Jann, Pfister, Franz MJ, Yuksel, Kamer Ali, Bischl, Bernd, Fietzek, Urban, Thomas, Janek
One major challenge in the medication of Parkinson's disease is that the severity of the disease, reflected in the patients' motor state, cannot be measured using accessible biomarkers. Therefore, we develop and examine a variety of statistical model
Externí odkaz:
http://arxiv.org/abs/1904.10829
Autor:
Sieberts, Solveig K., Borzymowski, Henryk, Guan, Yuanfang, Huang, Yidi, Matzner, Ayala, Page, Alex, Bar-Gad, Izhar, Beaulieu-Jones, Brett, El-Hanani, Yuval, Goschenhofer, Jann, Javidnia, Monica, Keller, Mark S., Li, Yan-Chak, Saqib, Mohammed, Smith, Greta, Stanescu, Ana, Venuto, Charles S., Zielinski, Robert, Glaab, Enrico, Jayaraman, Arun, Evers, Luc J. W., Foschini, Luca, Mariakakis, Alex, Pandey, Gaurav, Shawen, Nicholas, Synder, Phil, Omberg, Larsson
One of the promising opportunities of digital health is its potential to lead to more holistic understandings of diseases by interacting with the daily life of patients and through the collection of large amounts of real-world data. Validating and be
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od______2658::0de859d57a1235521622425cc1bb5be9
http://orbilu.uni.lu/handle/10993/54729
http://orbilu.uni.lu/handle/10993/54729
Publikováno v:
Big Data (2167-6461); Jun2023, Vol. 11 Issue 3, p181-198, 18p
Positive-unlabeled (PU) learning aims at learning a binary classifier from only positive and unlabeled training data. Recent approaches addressed this problem via cost-sensitive learning by developing unbiased loss functions, and their performance wa
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ec36a2f6c28fc5bba794b90d270647db
http://arxiv.org/abs/2201.13192
http://arxiv.org/abs/2201.13192