Completed
Proving the conjecture: Attempt 1
Class Central Classrooms beta
YouTube playlists curated by Class Central.
Classroom Contents
On the Hardness of Reinforcement Learning With Value-Function Approximation
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Reinforcement Learning (RL) Applications
- 3 Value-function Approximation
- 4 Comparison between SL and RL
- 5 Markov Decision Process (MDP)
- 6 Batch learning in MDPS
- 7 Example: Video game playing
- 8 Batch learning in large MDPS
- 9 Assumption on data (?)
- 10 Assumption on data & MDP dynamics
- 11 Algorithm for batch RL
- 12 How things go wrong (w/ restricted class)
- 13 Fix using a strong assumption ("completeness")
- 14 Realizability alone is insufficient?
- 15 Proving the conjecture: Attempt 1
- 16 Checklist for a plausible construction
- 17 Importance of the conjecture
- 18 Importance of the construction
