Deep Reinforcement Learning: Pong from Pixels
SMRTR summary
Reinforcement Learning (RL) is enabling computers to master complex tasks like playing ATARI games and Go through trial and error. Recent progress in RL has been driven primarily by advances in computing power, data availability, and infrastructure rather than fundamentally new algorithms. Policy Gradients has emerged as a preferred approach for many RL problems.
SMRTR provides this summary for quick context. The original article belongs to Hacker News.
Read the original article