A program that explains the systematic relationship between RL schedules and learned values of RL agents.
https://www.alignmentforum.org/posts/xqkGmfikqapbJ2YMj/shard-theory-an-overview
A program that explains the systematic relationship between RL schedules and learned values of RL agents.
https://www.alignmentforum.org/posts/xqkGmfikqapbJ2YMj/shard-theory-an-overview