This program implements Dyna-Q algorithm on blocking maze and shortcut maze games. This can be executed by simple run command interface in Colab. The plot for cumulative rewards for Dyna Q agent is shown for both Blocking maze and shortcut maze.
** Link to live notebook**
https://colab.research.google.com/drive/13V-efENYvWv-UAhYoo9RZOkBtiPiKBcj#scrollTo=hLEt8QUBKQ56