Scientific Data Services
Talk on Modified Nonlinear Least Squares Estimation and Inference - ZeSt Colloquium
Karsten Reichold from the Department of Statistics at the Technical University of Dortmund will give a talk on "Smooth Transition Cointegrating Regressions: Modified Nonlinear Least Squares Estimation and Inference" at the ZeSt Colloquium. The talk will take place on Tuesday, 13 June 2023 from 12:00 to 13:00 at the University of Bielefeld, room W9-109.
Talk on quantifying the labelling uncertainty for machine learning models - ZeSt Colloquium
As part of the ZeSt Colloquium Prof. Dr. Göran Kauermann from the Institute of Statistics at the Ludwig-Maximilians-Universität Munich will give his guest lecture "Categorizing the World into Local Climate Zones - Towards Quantifying Labeling Uncertainty for Machine Learning Models" at the University of Bielefeld on Tuesday, 30th of May. He will give an insight into the work on modelling labelling uncertainty in the context of remote sensing and satellite image classification. The talk will be focused on the construction of a multinomial mixed model based on the assumption that there is no ambiguity in the image classes, but apparently only in the experts' opinions about them.