Learn in One : Reinforcement Learning basics [ Part 1 ] |
Welcome to "Learn in One" Today, we plunge into the world of Reinforcement Learning.
Imagine training a dog, where it learns to perform tricks based on rewards or punishments. Reinforcement Learning operates on a similar principle, where a machine or software agent learns to make decisions by performing certain actions in an environment to achieve a goal.
As the decision-maker, the agent learns a policy, which is a strategy dictating the agent's subsequent action based on the current circumstance, akin to a roadmap.
The complete spectrum of prospective actions an agent can execute is known as the action space. The agent's goal is to traverse this action space to boost its reward.
Let's dive into some crucial terms. The State-Action-Reward-State-Action (SARSA) is a method for learning a Markov Decision Process policy. To illustrate SARSA,
Imagine a chess game where the agent learns from its mistakes and successes, adapting its future actions based on the rewards or punishments received for its moves.(short pause)