The AI Alignment Forum

FEBRUARY 28, 2025

Some notable possible candidates include: Goodharts Law. Which specification learning algorithms are guaranteed to converge to a good specification? How do these errors depend on how much optimisation pressure we exert, and other relevant parameters? Are there any distinct failure modes that could be individuated and characterised?

Research 40

Research Learning Method Policy 40

The AI Alignment Forum

FEBRUARY 28, 2025

Goodhart's Law in Reinforcement Learning As you probably know, "Goodhart's Law" is an informal principle which says that "if a proxy is used as a target, it will cease to be a good proxy". This paper is also discussed in this post (Paper 4). For details, see the full paper.

Learning 40

Learning Discussion Classes Policy 40

Google Research AI blog

NOVEMBER 28, 2022

A Workshop for Algorithmic Efficiency in Practical Neural Network Training Workshop Organizers include: Zachary Nado , George Dahl , Naman Agarwal , Aakanksha Chowdhery Invited Speakers include: Aakanksha Chowdhery , Priya Goyal Human in the Loop Learning (HiLL) Workshop Organizers include: Fisher Yu, Vittorio Ferrari Invited Speakers include: Dorsa (..)

Google 52

Google Language Tutorial Offline 52

The AI Alignment Forum

MARCH 28, 2025

Daniel Filan (00:28:50): If people remember my singular learning theory episodes , theyll get mad at you for saying that quadratics are all there is, but its a decent approximation. (00:28:56): Or is that Daniel Filan (01:40:07): Well, if we knew the Chinchilla scaling law , we would know this. Jason Gross (01:40:16): Probably.

Model 52

Model Network Train Training 52