Completed
[00:00] - Introduction to building reliable agents with RL.
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
How to Train Your Agent - Building Reliable Agents with RL
Automatically move to the next video in the Classroom when playback concludes
- 1 [00:00] - Introduction to building reliable agents with RL.
- 2 [00:49] - Case Study: ART-E, an AI email assistant.
- 3 [02:19] - The importance of starting with prompted models before moving to RL.
- 4 [03:17] - Performance improvements of RL over prompted models.
- 5 [05:18] - Cost and latency benefits of the RL approach.
- 6 [08:02] - The two hardest problems in modern RL: realistic environments and reward functions.
- 7 [13:13] - Optimizing agent behavior with "extra rewards."
- 8 [15:25] - The problem of "reward hacking" and how to address it.
- 9 [18:37] - The solution to reward hacking:
