Detail of the student project

List
Topic:Adaptace klasifikátoru na změnu apriorních pravděpodobností
Department:Katedra kybernetiky
Supervisor:Ing. Milan Šulc
Announce as:Diplomová práce, Semestrální projekt
Description:Výzkum metod pro adaptaci klasifikátorů na změnu apriorních pravděpodobností tříd (prior shift, label shift) mezi trénovacími a testovacími daty.
Výstupem bude technická zpráva a GIT repozitář.
Bibliography:1] Amr Alexandari, Anshul Kundaje, and Avanti Shrikumar. Maximum likelihood with bias-corrected calibration is hard-to-beat at label shift adaptation. ArXiv, 1901.06852v5, 2019.1
[2] Kamyar Azizzadenesheli, Anqi Liu, Fanny Yang, and Animashree Anandkumar. Regularized learning for domain adaptation under label shifts. arXiv preprint arXiv:1903.09734,2019. 1, 2
[3] Marthinus Christoffel du Plessis and Masashi Sugiyama. Semi-supervised learning of class balance under class-prior change by distribution matching. CoRR, abs/1206.4677, 2012.1, 2, 4
[4] Zachary C. Lipton, Yu-Xiang Wang, and Alex Smola. Detecting and correcting for label shift with black box predictors, 2018. 1, 2, 4
[5] Geoffrey J. McLachlan. Discriminant Analysis and Statistical Pattern Recognition. John Wiley & Sons, Inc, Hoboken, NJ,USA, 1992-03-27. 2
[6] Marco Saerens, Patrice Latinne, and Christine Decaestecker. Adjusting the outputs of a classifier to new a priori probabilities: A simple procedure. Neural Comput., 14(1):21–41, Jan.2002. 1, 2, 4
[7] Milan Sulc and Jiri Matas. Improving CNNclassifiers by estimating test-time priors. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, Oct 2019. 1, 2, 4
[8] Slobodan Vucetic and Zoran Obradovic. Classification on data with biased class distribution. In European Conference on Machine Learning, pages 527–538. Springer, 2001. 1
Responsible person: Petr Pošík