CS 285 at UC Berkeley

Deep Reinforcement Learning

Lectures: Mon/Wed 5:30-7 p.m., Online

Are you a UC Berkeley undergraduate interested in enrollment in Fall 2021? Please do not email Prof. Levine about enrollment codes. We will post a form in August 2021 where you can fill in your information, and students will be notified after the first week of class. PhD students will be able to enroll into the class directly as normal.

Lectures will be recorded and provided before the lecture slot. The lecture slot will consist of discussions on the course content covered in the lecture videos.

Piazza is the preferred platform to communicate with the instructors. However, if for some reason you wish to contact the course staff by email, use the following email address: cs285fall2020@googlegroups.com.

Lecture recordings from the current (Fall 2020) offering of the course: watch here
Enrolled students: please use the private link you were provided, not this one!

Looking for deep RL course materials from past years?

Recordings of lectures from fall 2019 are here, and materials from previous offerings are here.
  • Instructor Sergey Levine


    Office Hours: After lecture

  • Head GSI Michael Janner


    Office Hours: Tuesday 3-4 pm

  • GSI Vitchyr Pong


    Office Hours: Friday 1-2pm

  • GSI Aviral Kumar


    Office Hours: Thursday 2-3pm

  • uGSI Alexander Khazatsky


    Office Hours: Monday 12-1pm

Week 1 Overview

Course Introduction, Imitation Learning

Week 2 Overview

Imitation Learning

Week 4 Overview

Policy Gradients and Actor Critic

Week 5 Overview

Value Functions and Q-learning

Week 6 Overview

Advanced Policy Gradients and Model-based learning

Week 7 Overview

Advanced Model Learning and Imitating Optimal Controllers

Week 9 Overview

Offline Reinforcement Learning and RL Theory

Week 10 Overview

RL Algorithm Design and Variational Inference

Week 11 Overview

Control as Inference and Inverse Reinforcement Learning

Week 12 Overview

Transfer Learning and Multi-Task Learning

Week 13 Overview

Meta-Learning, Challenges, Open Problems