Reinforcement Learning Basics Theory
Policy, Reward, MDP, Monte-Carlo, Temporal Difference, etc... These are concepts that are often encountered in reinforcement learning-related papers or projects, but there doesn't seem to be many lecture materials that explain the exact definition of each term from the very basics. If you jump into a paper or project without a solid understanding of these, you will end up drifting in the vast ocean without a direction, like a ship that has lost its rudder.
I think the best resource that explains the basics with rich explanations and intuitive understanding is Professor D. Silver of DeepMind's YouTube lecture. However, the lecture is conducted in English and is somewhat difficult for beginners to listen to, so this lecture aims to re-convey the same content in a more understandable way in Korean. Just as D. Silver's lecture is composed of 10 lectures, our lecture will also be composed of 10 lectures.
Helpful people
- Those who want to solidify the theory and basic concepts of reinforcement learning
- Anyone who wants to learn how deep learning is applied to reinforcement learning
AlphaGo paper review
If you are curious about what you can do by learning reinforcement learning, please first watch our Pangyo Lab's AlphaGo paper review video.
AlphaGo paper review:
https://www.youtube.com/watch?v=SRVx2DFu_tY&list=PLpRS2w0xWHTfnWmr95LtIu4v4HbVxqTlM AlphaGo Zero Paper Review:
https://youtu.be/CgOGKChwWrw
What is reinforcement learning?
Reinforcement Learning, one of the fields of machine learning
Machine learning can be broadly divided into supervised learning, unsupervised learning, and reinforcement learning. It is a method of recognizing the current state and selecting an action or action sequence that maximizes reward among selectable actions.
Introducing the knowledge sharer
The old monk
Seoul National University - Computer Engineering, Economics (2010-2015)
Seoul National University Graduate School of Convergence Science and Technology - Research on Hyperparameter Optimization of Deep Learning (2015-2017)
NCSoft AI Research - Artificial Intelligence Researcher, Reinforcement Learning Team (2017-)
Jeon Min-young
Seoul National University - Computer Engineering, Visual Design (2011-2017)
Gameberry - Developer (2014)
Ringle - Developer (2015)
Madup - Developer (2016-2017)
Naver - Papago Team Front-end Development (2018-)