Basic Formalisms of Reinforcement Learning
If you are interested and want to start learning about Reinforcement Learning it is important for you to know the key concepts and formalisms. In this article I want to cover the basic concepts of Reinforcement Learning like:
- What is an Agent
- What is an Environment
- What is the Reward / Reward-function
- What is the Action-space and State-space
The goal of this article is to make a short introduction to these terms and show how they interlock with each other.
Recap
Reinforcement learning is inspired by the human learning and decision-making process of “trial and error”. As a member of the machine learning family it is the only one with no previously collected data. Therefore it has to collect its data or experience along the learning progress. But how does this happen and what are the mechanisms behind that…
(for further reading and a quick introduction to where Reinforcement Learning is set up in the field of AI check my previous article!)
