Due to the short cycle, high gold content, and good learning atmosphere of Baidu AI training camp, I have been participating in their training camp. The theme of this issue is intensive learning. I have been exposed to machine learning and deep learning before, but I haven’t learned about reinforcement learning. This time I have the opportunity to learn about it. The course link is as follows: https://aistudio.baidu.com/aistudio/education/group/info/1335
1. What is reinforcement learning
So in simple terms, reinforcement learning belongs to machine learning, but it is more about responding to the environment.
2. What can reinforcement learning do?
3. The difference between reinforcement learning and supervised learning
4. How does reinforcement learning solve problems
Five, reinforcement learning algorithm and environment
The above is a brief introduction to reinforcement learning, which is also the main topic of these seven days. We have used Q-learning, Sarsa, DQN, Policy Gradient and other algorithms to solve the optimal strategy problem of some small games, and practiced the PARL reinforcement learning framework. I have to say that this framework is really easy to use, and it has a lot of packaging. The environment and algorithms are very friendly for beginners.
The above is the 7-day check-in camp of this issue. The world champion lady led us into intensive learning with zero foundation, and gained a lot. Looking forward to the next check-in camp.