Hi, today the research in Deep Reinforcement Learning is more about Actor Critic Algorithms and PPO. But you need to understand DQN and Policy Gradients before working on Actor Critic and PPO.

If you want to learn more about research in this topic look at DeepMind and OpenAI websites.

I will write an article about A2C that will be published next week and a one about PPO in 2 weeks.

Yes if you have some questions about your code I can look at it, I’m not the most skilled guy but I can try ^^.