Opening the Black Box of Machine Learning

Type of thesis: Bachelor- und Masterarbeit / location: Dresden / Status of thesis: Open theses

The thesis aims to make the ML prediction interpretable. The main focus is studying Variable Importance (VI), Partial Dependence Plots (PDP), etc. Recent literature proposes different techniques for their computation. Your task is to compare them for selected ML methods on several datasets.

Counterpart

Dr. Iryna Okhrin

TU Dresden

Machine Learning, Data Analytics, Statistics, Living Lab

TU
Universität
Max
Leibnitz-Institut
Helmholtz
Hemholtz
Institut
Fraunhofer-Institut
Fraunhofer-Institut
Max-Planck-Institut
Institute
Max-Plank-Institut