202509281152 Status: #idea Tags: #reinforcement_learning #ai # The environment in RL is everything not contained in the agent The agent is the decision maker in an RL problem. For instance, suppose we we're training a robot to pick up objects *Only* the code is part of the agent. The robot arm, the objects it's interacting with, and everything else are *outside* the agent and hence part of the environment. "The decision maker, the agent, can only have a single role: making decisions. Everything that comes after the decision gets bundled into the environment." --- # References [[Grokking Deep Reinforcement Learning]]