Reinforcement learning core theory that is easy to understand

Why You Should Learn Reinforcement Learning Now 😆

Simply put, reinforcement learning refers to a computer program learning what actions to take in a given situation . It can be understood as a type of machine learning that can easily solve difficult decision-making problems. Reinforcement learning is one of the most essential aspects of AI decision-making, enabling machines to design programs that interact with each other and reach optimal conclusions.

But it's hard. It's hard. It's too hard 😱

Despite our desire to study reinforcement learning, studying it alone is incredibly difficult. Here are two major reasons:

• It is difficult to understand the principles contained in representative reinforcement learning formulas.
• A variety of concepts are mixed in one algorithm, such as TD vs MC, Value-based vs Policy-based, etc.

Even if you have a solid foundation, your learning speed will be boosted 🔥

If you're studying reinforcement learning, you might dream of cutting-edge papers and dazzling demos, but for a more distant future, you need to thoroughly master the fundamentals of reinforcement learning. This course will provide you with a deep understanding of the fundamental concepts of reinforcement learning and help you achieve the following:

What's special about my lecture 😗

• Just pick out each core concept of reinforcement learning!
  Multi-armed bandits, Markov decision processes, dynamic programming, Monte Carlo methods, etc.
  We will teach you the core concepts necessary for reinforcement learning from the basics.
  
  
• Understand which reinforcement learning techniques to use for a specific problem!
  It doesn't end with just knowing the theory. You will experience the actual practice process with me.
  You will be able to intuitively understand how to respond in certain situations.
  
  
• Embody concepts through various practices!
  After learning the basic theory, you can immediately apply what you have learned through practice.
  Let me help you make it yours.

What you will learn in this lecture 😝

Reinforcement Learning Basics

Learn about the basic theory and framework of reinforcement learning, and explore basic concepts related to AI reward systems, such as multi-armed bandits, Markov states, rewards, and state transitions.

dynamic programming

Based on the Markov decision process we learned earlier, we will learn about the definition and application of dynamic programming.

Monte Carlo method

Learn the basic concepts of the Monte Carlo method, which is used when the values you want to calculate are complex.

Time-lapse learning

We'll explore the fundamental concepts of temporal learning, a form of learning directly from real-world learning experiences, compare it directly with the Monte Carlo method, and see in what situations it can be utilized.

Model-based reinforcement learning

Learn about the concept of models in machine learning, explore reinforcement learning theory based on tables and models, and practice it yourself.

Policy-based reinforcement learning

Learn about policy-based reinforcement learning, which selects actions based on state rather than determining actions based on a value function.

I am this kind of person 😝

Before attending the lecture, please check any questions you may have in advance! 😝

By any chance, have you taken this course? ✨

Machine Learning Lecture by Chris Song, Knowledge Sharer

1. Machine Learning Engineer Practice