huixxi / cs234-reinforcement-learning-winter-2019

My Solutions of Assignments of CS234: Reinforcement Learning Winter 2019

CS234-Reinforcement-Learning

My Solutions of Assignments of CS234: Reinforcement Learning Winter 2019

---

Lecture Videos

This course contains 15 lecture videos, and you can watch them from youtube and bilibili(vpn free).

• YouTube StanfordOnline
• Bilibili

You can download the slides for each lecture from here.

---

Assignments

• assignment 1

• assignment 2

• assignment 3

---

Highly Recommended RL Resources

• Resources List

  • RL Resources

• TextBooks

  • Reinforcement Learning: An Introduction 2nd Edition. (R.Sutton, G.Barto)
  • Deep Learning. (Yoshua Bengio)

• RL Courses

  • OpenAI Spining Up
  • RL-DeepMind Tutorials 2018 (Video)

• Papers:
  Deep Q Learning

  • DeepMind Nature Paper
  • Playing Atari with Deep Reinforcement Learning
  • Prioritized Experience Replay

  Imitation Learning

  • Generative Adversarial Imitation Learning(GAIL)
  • Maximum Entropy Inverse Reinforcement Learning

• Blogs:

  • Policy Gradient Algorithm
  • Meta Reinforcement Learning
  • Rainbow

• Podcasts:

  • TalkRL
  • DeepMind: The Podcast

---

Author

@Huxixi

cs234-reinforcement-learning-winter-2019's Issues

training does not work

Hey there,
I am training the q5 problem, but the training is not going well. I wondered if you have any clue what is wrong.

Here is a copy of training data:
122501/5000000 [..............................] - ETA: 34518s - Loss: 0.1562 - Avg_R: -21.0000 - Max_R: -21.0000 - eps: -55.3107 - Grads: 0.0000 - Max_Q: 0.0000 - lr: -0.0048

The gradient is 0, Max_Q is 0. Reward has not improved at all.

Do you have the "official solutions"?

Until today, the solutions were on the course website but apparently they have taken down everything.
Would be nice if you could also upload them in the repo or attach them here
Thanks in advace!