Winter School 1B - Munich

• Probability theory basics 

• Theoretical distributions 

• Visualization techniques 

• Statistical hypothesis testing 

• Multiple testing problem and correction methods 

• Basic topics from unsupervised learning 

• Performance estimation via resampling techniques  

• Important performance metrics for regression and classification, and ROC analysis basics 

• Data Preprocessing and Imbalanced Learning 

Dinner and “Eisstockschießen” / Bavarian curling   

• From a single neuron to single-hidden-layer networks  

• Multi-layer feedforward networks  

• Training neural networks   

• Regularization: basics and for nonlinear models; weight decay vs L2 

• Multiple testing problem and correction methods 

• Convolutional neural networks: intuition and building blocks.