Home

Deep RL

Reinforcement learning considers the problem of learning to act and is poised to power next generation AI systems, which will need to go beyond input-output pattern recognition (as has sufficed for speech, vision, machine translation) but will have to generate intelligent behavior. Example application domains include robotics, marketing, dialogue, HVAC, optimizing healthcare and supply chains.

Reinforcement learning poses significant challenges beyond pattern recognition, including exploration, credit assignment, stability, safety. While these challenges are far from solved, there have recently been several major success stories. This includes learning to play Atari games from raw pixels, beating the Go World Champion, learning complex locomotion behaviors, acquiring advanced manipulation skills, and controlling datacenter energy consumption. These successes have relied on the synergy between deep neural nets and reinforcement learning, i.e., deep reinforcement learning (Deep RL).

​This two-day long bootcamp will teach you the foundations of Deep RL through a mixture of lectures and hands-on lab sessions, so you can go on and build new fascinating applications using these techniques and maybe even push the algorithmic frontier.

​Assumptions on prior knowledge will be tuned to what we find out about the audience in the applications.

The Bootcamp

We will cover the following topics:

• RL Basics
• Policy Gradients
• Actor-Critic Algorithms
• Q-learning
• Evolution Strategies
• RL trouble-shooting and debugging strategies
• Current research frontiers

There will be tutorials sessions for all topics, followed by hands-on lab sessions using OpenAI Gym.

​Learning Goals

• Understanding the foundations
• Ability to implement state-of-the-art methods from scratch
• Ability to build advanced applications on top of rllab
• Ability to apply Deep RL to new domains

Organizers

Guest Instructors

Teaching Assistants

• Joshua Achiam (OpenAI and UC Berkeley)
• Marcin Andrychowicz (OpenAI)
• Richard Chen (OpenAI)
• Ignasi Clavera (UC Berkeley)
• Carlos Florensa (UC Berkeley)
• Tuomas Haarnoja (UC Berkeley)
• Rein Houthooft (OpenAI)
• Sandy Huang (UC Berkeley)
• Thanard Kurutach (UC Berkeley)
• Yang Liu (OpenAI and UIUC)
• Andrew Liu (UC Berkeley)
• Rohin Shah (UC Berkeley)
• Adam Stooke (UC Berkeley)
• Haoran Tang (UC Berkeley)
• Jie Tang (OpenAI)
• Garrett Thomas (UC Berkeley)
• Tianhe Yu (UC Berkeley)
• Marvin Zhang (UC Berkeley)
• Tianhao Zhang (UC Berkeley)

Sponsors

Diamond

Gold

Silver

Apply Now

Update: We are no longer accepting applications, but if you'd like to get an email update if/when there might be a second edition, please fill in your information here.

​If you have any questions, feel free to contact us at [email protected].