I have a situation where I'm trying to maximize the number of steps in a fixed training time frame. It's possible that specific steps will lead to a delay until the agent can act again, and thus less overall steps by the end of the fixed length episode. I am using a basic DQN.
I want to try to explain to a layman how having a constant reward for every step, if given a fixed time frame during training, can allow the agent to try to maximize the number of steps within the episode. Conceptually it makes sense to me, but I'm at a bit of loss on how to explain it.
I seem to be missing the right keywords to be searching for examples of this situation online. Any help with keywords and/or explanation would be helpful.