Skip to content

Comment on The Great AI Weirding

Comments

What is RL?

Reinforcement Learning. They are referencing a concept known as Reward Hacking (see Robert Miles videos for a high level explanation). You may be familiar with the concept already though, see Goodhart's Law.

Reinforcement learning

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.