What I was looking for is: A DQN (Deep Q Learning Neural Network) or a Reinforcement Learning example that can learn from existing simulation data, and then can use that learning to interactively optimize an objective. The challenge will be: Whether my data can be learned from (whether the format/structure of the data is usable in DQN/RL) by the DQN/RL, also what to define as the actions, and how to define, utilize, and optimize the reward. Came across misc. stuff as below:
Came across: Did not really check: Reinforcement Learning – A Simple Python Example and A Step Closer to AI with Assisted Q-Learning
https://www.youtube.com/watch?v=nSxaG_Kjw_w&index=1&list=UUq4pm1i_VZqxKVVOz5qRBIA
The above might have used the following:
https://amunategui.github.io/reinforcement-learning/index.html
A Hands-On Introduction to Deep Q-Learning using OpenAI Gym in Python
https://www.analyticsvidhya.com/blog/2019/04/introduction-deep-q-learning-python/
Top 7 Python Libraries For Reinforcement Learning
https://analyticsindiamag.com/python-libraries-reinforcement-learning-dqn-rl-ai/
*** . *** *** . *** . *** . ***
Courses: http://Training.SitesTree.com (Big Data, Cloud, Security, Machine Learning)
Blog: http://Bangla.SaLearningSchool.com, http://SitesTree.com
8112223 Canada Inc./JustEtc: http://JustEtc.net
Shop Online: https://www.ShopForSoul.com/
Linkedin: https://ca.linkedin.com/in/sayedjustetc
