Course Preview

Deep Reinforcement Learning and Control
Instructor: Katerina Fragkiadaki, Ruslan Satakhutdinov
Department: Machine Learning
Institution: Carnegie Mellon University
Platform: Independent
Price: Free
Prerequisites: reinforcement learning, numerical optimization, machine learning

Suggested relevant courses in MLD are 10701 Introduction to Machine Learning, 10807 Topics in Deep Learning, 10725 Convex Optimization, or online equivalent versions of these courses. Students less familiar with reinforcement learning can warm start with the first chapters of Sutton&Barto and with the first lectures of Dave Silver’s course.

Richard S. Sutton and Andrew G. Barto. Reinforcement Learning: An Introduction. 1998
Ian Goodfellow and Yoshua Bengio and Aaron Courville. Deep Learning. 2016.
Class Goal: 1. Implement and experiment with existing algorithms for learning control policies guided by reinforcement, expert demonstrations or self-trials. 2. Evaluate the sample complexity, generalization and generality of these algorithms. 3. Be able to understand research papers in the field of robotic learning. 4. Try out some ideas/extensions of your own. Particular focus on incorporating true sensory signal from vision or tactile sensing, and exploring the synergy between learning from simulation versus learning from real experience.