Beginner · Reinforcement Learning

State / Action

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. The two core elements of every reinforcement-learning step: what the world looks like and what the agent does.

Technical Definition

The two core elements of every reinforcement-learning step: what the world looks like and what the agent does.

How it works

A state s captures all information the agent needs to decide what to do next; an action a is one of the choices available in that state. The environment responds with a new state and a reward. The state space and action space can be discrete (chess moves) or continuous (robot joint angles), which shapes which RL algorithms apply.

Related Concepts

• Reinforcement Learning — A paradigm where an agent learns to make decisions by receiving rewards or penalties from its environment through trial and error.
• Policy — In reinforcement learning, the strategy an agent follows to choose actions given a state.
• Reward Function — A scalar signal that tells a reinforcement-learning agent how good its action was.