Course Preview

Tutorial on Deep Reinforcement Learning
Instructor: John Schulman
Institution: Machine Learning Summer School
Platform: Independent
Year: 2016
Price: Free
Description:
This tutorial will cover deep reinforcement learning, with 4 recored video lectures. Each video is around one-hour long. The lecture slide and lab material are available at http://learning.mpi-sws.org/mlss2016/speakers/ at "deep reinforcement learning" section taught by John Schulman from UC Berkeley. Reinforcement learning studies decision making and control, and how a decision-making agent can learn to act optimally in a previously unknown environment. Deep reinforcement learning studies how neural networks can be used in reinforcement learning algorithms, making it possible to learn the mapping from raw sensory inputs to raw motor outputs, removing the need to hand-engineer this pipeline. The aim of the tutorial is to introduce you to the most important techniques in deep reinforcement learning. This course will include hands-on labs, where you will implement the algorithms discussed in the lectures. For the labs, you should have a working installation of OpenAI Gym. A Python-based autodiff library such as Theano or Tensorflow is highly recommended.