News for hackers. Stuff that matters.
Re:Arc Prize 2024 Winners and Technical Report by YeGoblynQueenne on Saturday December 07, 2024 @09:13PM (#42354492)
That's fine if you want to measure sample efficiency, but ARC-AGI is supposed to measure progress towards AGI.
Re:Arc Prize 2024 Winners and Technical Report by trott on Saturday December 07, 2024 @10:04PM (#42354728)
> That's fine if you want to measure sample efficiency, but ARC-AGI is supposed to measure progress towards AGI.

On the Measure of Intelligence defines intelligence as skill-acquisition efficiency, I believe, where efficiency is with respect to whatever is the limiting factor. For each ARC task, the primary limiting factor is the number of samples in it. And the skill here is your ability to convert inputs into the correct outputs. In other words, in this context, intelligence is sample-efficiency, as I see it.

Re:Arc Prize 2024 Winners and Technical Report by YeGoblynQueenne on Sunday December 08, 2024 @07:06AM (#42356669)
Is that what fchollet is claiming?
Re:Arc Prize 2024 Winners and Technical Report by trott on Sunday December 08, 2024 @11:38AM (#42358052)
Not sure. But I think this follows logically from the definition of intelligence he is using. Also, see II.2.2 in the paper.