3 ms·
Reinforcement Learning
by pmalynin 9y ago
Reinforcement Learning
- gugagore 9y agoI think really just "Value Iteration" (which isn't just used in RL). Reinforcement Learning itself is a problem setting and there are solutions in RL that don't use dynamic programming (for example, policy gradient methods).
- pmalynin 9y agoWell of course :) but it’s a cute method to solve certain MDPs