Current, content updated Oct 2026
Deep Reinforcement Learning
Udacity / Nvidia Deep Learning Institute on Udacity
Udacity's advanced nanodegree on deep reinforcement learning: DQNs, policy gradients, actor-critic methods and multi-agent RL, with hands-on projects.
Our take
Still one of the most thorough practical RL programmes, with well-known projects like the tennis-playing agents. Advanced; about 83 hours.
Best for: Engineers comfortable with Python and deep learning who want to go deep on RL.
Based on the course page, syllabus and pricing, not on completing the course. How we rank
Tools and models covered
- PyTorch
Based on the course content as of 2026-10.