not much information is given regarding that so far, i was curious about that too
rokosbasilisk
Karma: 30
Hutter-Prize for Prompts
“Algorithm for Concept Extrapolation”
Regarding the first one i am not expecting a single-prompt to generate the entirity of enwiki8/9. I am more interested in finding a set of prompts with a lookup table if possible to replicate enwiki data.
Thanks for the pointer for chincilla post, will look into it.
The AGI needs to be honest
Goal-misgeneralization is ELK-hard
though as Geoff Hinton has pointed out, ‘confabulations’ might be a better word
I think yann lecun was the first one to using this word https://twitter.com/ylecun/status/1667272618825723909
I don’t see any recent publications for paul christiano related to this. So i guess the problem[s] is still open.
parameters before L is less than ,
should this be after?
47 layers layer
47 layers later ?
really interesting idea.
not yet
This requires hitting a window—our data needs to be good enough that the system can tell it should use human values as a proxy, but bad enough that the system can’t figure out the specifics of the data-collection process enough to model it directly. This window may not even exist.
are there any real world examples of this? not necessarily in human-values setting
From a complexity theoretic viewpoint, how hard could ELK be? is there any evidence that ELK is decidable?
is there a separate post for “train a reporter that is useful for another AI” proposal?
silly idea: instead of thought-annotating ai-dungeon plays, we can start with annotating thoughts for akinator gameruns.
pros: much more easier and faster way to build a dataset, with less ambiguity
cons: somewhat restricted than the original idea.
The proof need not be bogus, it can be a long valid proof but since you are describing the problem in natural language, the proof generated by the AGI need not be for the problem that you described.
Also, the AGI can generate a long valid proof but it may not be for the question you have asked, since the assumption is that the problems described in natural language and its the AGI’s job to understand and convert it to formal language and then prove it
I think instead of recursively asking for higher level proof it should be a machine-checkable regarding the correctness of the AGI itself?
verifying a proof may run in polynomial time compared to the exponential of finding one, but it doesn’t rule out the possibility that there exist a large enough proof which will be hard to check.
There are many algorithms which are polynomial in time but are far worse to run in reality.
It looks like you are confusing autoGPT with babyagi which was created by yohei nakajima who is a VC. the creator of autoGPT (Toran Bruce Richards) is a game-developer with a decent programming (game-development) experience. Even the figure shown here is that from babyagi (https://yoheinakajima.com/task-driven-autonomous-agent-utilizing-gpt-4-pinecone-and-langchain-for-diverse-applications/).