# Reinforcement learning, is normalizing the reward should also change the learning rate?

The reward range for my current environment is $$[-10000, 5000]$$, so currently I normalize the reward to $$[-2, 1]$$. Should I also change the learning rate?