The Lunar Lander DQNAgent is again the same, with the exception of a much smaller learning rate, as you can see from the following code:
dqn = DQNAgent(model=model, nb_actions=num_actions, memory=memory, nb_steps_warmup=10, target_model_update=1e-2, policy=policy)
dqn.compile(Adam(lr=0.00025), metrics=['mae'])