Policy search
The explanation is being written.
Sources and connections are already here. The explanation will say what it is, what it is for and how to do it.
Timeline
- 1992Williams: simple statistical gradient-following algorithms for connectionist reinforcement learning. source