📇
DeepRL
  • CS285: Deep RL Notes
  • RL Fundamentals
  • Policy Gradient
    • Policy Gradient Basics
    • Actor Critic Algorithms
    • Advanced Policy Gradients
  • Value Based Methods
    • Policy and Value Iteration Algorithms
    • DQN and beyond
  • Model-based Methods
    • Model-based Planning and Model-based Predictive Control
    • Model-based Policy Learning
  • Inference, Control, and Inverse RL
    • Latent Models and Variational Inference
    • Control as Inference
    • Inverse Reinforcement Learning
  • Transfer Learning in RL
    • Transfer and Multi-task Learning
    • Paper Reading Notes
  • Coming soon...
    • Offline RL
    • RL from Pixels
Powered by GitBook
On this page

CS285: Deep RL Notes

For UC Berkeley CS285: Deep Reinforcement Learning, Decision Making, and Control, taught by Professor Sergey Levine.

Official Course Website: http://rail.eecs.berkeley.edu/deeprlcourse

These are my personal notes and wordy explanations on the core topics covered in this course, it’s meant to be a reference and sanity check for myself and for others learning deep RL. I typically write extensively about the most important parts of each lecture, so the notes are not (yet) comprehensive. I highly recommend watching the recorded lectures if you are self-studying for this class, professor Levine is a fantastic lecturer and many student questions that got recorded were also really interesting.

All screenshots/images in these notes credit to CS285 lecture slides. I'll try to keep updating new topics whenever possible, please reach out to me to mandi.zhao@berkeley.edu if you’d like to contribute to writing or beautifying these notes!

Happy Reinforcement Learning! :)

NextRL Fundamentals

Last updated 4 years ago