This repository includes the research matierals, proposal and implementation of a comparative analysis of reinforcement learning algorithms.
This folder includes the self thinking of the research goal of Forschungspraxis.
The implementations of GQ, RO-GQ and OSK-Q are in this directory. In the subfolder data, you can find the analytical data and its visualization of the project from the perspective of convergence, running time and sparsity.
Here is the report for the whole Forschungspraxis.