Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Can LLMs invent better ways to train LLMs?

sakana.ai

Post reply on HN
Print view View original

21–30 of 39 posts

Re: Can LLMs invent better ways to train LLMs?

#21

Post by 2Gkashmiri »

qingcharles wrote: post #18
IshKebab wrote: post #16

Earlier quoted context omitted.

I think it's probably unreasonable to expect them to without giving them the ability to experiment and test ideas. Pretty much no inventions were invented just by thinking, which is the environment most LLMs have.

I think this is the key point. There is no feedback loop right now.

is it possible for copilot or say llama or gpt4o to suggest a piece of code and actually go and try to run a test that they design on an ide and see if there are any results and try to fix issues?

right now you ask llm to write a code to do basic web scraping for HN website for latest url and give username of the submitter. sure they will give you a code and give you a test script but you as the user have to run the script and give manual feedback to LLM. if the testing step can be automated, user would give an input and desired output or a prompt and choose between the results, that would be good.

kinda like you do inpainting and outpainting and other painting stuff but for code.

feedback is the key here as you said

Re: Can LLMs invent better ways to train LLMs?

#22

Post by vidarh »

2Gkashmiri wrote: post #21
qingcharles wrote: post #18

Earlier quoted context omitted.

I think this is the key point. There is no feedback loop right now.

is it possible for copilot or say llama or gpt4o to suggest a piece of code and actually go and try to run a test that they design on an ide and see if there are any results and try to fix issues? right now you ask llm to write a code to do basic web scraping for HN website for latest url and give username of the submitter. sure they will give you a code and give you a test script but you as the user have to run the…

Genetic Programming was a thing in the 90s but hampered by a combination of the inefficiency of largely random mutations (plus some crossover, which was still largely undirected) with low odds of doing anything helpful, and lack of computational speed to test. A GP framework attempting to use LLMs to apply more or less "reasoned" changes within the same structure of generations of "mutations" tested against each other and previous generations best would be interesting.

Re: Can LLMs invent better ways to train LLMs?

#23

Post by rpigab »

dietr1ch wrote: post #3
bugbuddy wrote: post #2

Betteridge's law of headlines: no

You'll need the following law too, No clickbaity article is worth reading.

I hate clickbait, but this doesn't look like this one is, because it's not misleading, it's not trying to hide something to get you to click, and does not lure you with information that's not in the article.

Re: Can LLMs invent better ways to train LLMs?

#24

Post by steve1977 »

mdp2021 wrote: post #8
ChuckMcM wrote: post #5

A better question is "Can LLMs invent anything ?" Don't misunderstand, building systems models using existing system response as a way of analyzing those systems is a useful methodology and it makes some things otherwise tedious things not so tedious. Much like "high level" languages removed the tedium of writing in assembly code. But for the same reason that a compiler won't emit a new, more powerful, CPU instructio…

> Can LLMs invent anything Can they propose a working novelty: -- after deep thought about idea soundness, probably not at this stage -- through cycles of trials, not knowing exactly why - probably yes After all, your hammer needs not intelligence.

But my hammer is totally useless without me as a human using it and telling it exactly what to do.

Re: Can LLMs invent better ways to train LLMs?

#25

Post by steve1977 »

paulddraper wrote: post #9
ChuckMcM wrote: post #5

A better question is "Can LLMs invent anything ?" Don't misunderstand, building systems models using existing system response as a way of analyzing those systems is a useful methodology and it makes some things otherwise tedious things not so tedious. Much like "high level" languages removed the tedium of writing in assembly code. But for the same reason that a compiler won't emit a new, more powerful, CPU instructio…

What are inventions other than combinations of existing patterns? What is the human mind if not a computer? What is the universe if not repeated regurgitations of the four fundamental forces and 12 particles?

I guess the key aspect for human inventions is a stochastic element to the combination of existing pattern. I.e. seeing (or imagining) connections that are not obvious.

Re: Can LLMs invent better ways to train LLMs?

#27

Post by luke-stanley »

The project sounds quite interesting but I'm not sure running it is going to work! The code `gpt_model = "gpt4_20231230_1106preview"` is not using a valid model name as best as I can tell, so it seems unlikely to work - from https://github.com/SakanaAI/DiscoPOP/blob/main/scripts/launc... Unusually, the issue section doesn't exist so I can't provide feedback to them that way. But luchris429's repo does have it so will do so there. Maybe it's dead code. Still, it's wrong.

Re: Can LLMs invent better ways to train LLMs?

#28

Post by binary132 »

blowski wrote: post #4
dietr1ch wrote: post #3

Earlier quoted context omitted.

You'll need the following law too, No clickbaity article is worth reading.

Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something. (From the guidelines)

Fine then: “don’t write articles with question headlines or people can safely assume the answer is no.”

Re: Can LLMs invent better ways to train LLMs?

#29

Post by luchris429 »

luke-stanley wrote: post #27

The project sounds quite interesting but I'm not sure running it is going to work! The code `gpt_model = "gpt4_20231230_1106preview"` is not using a valid model name as best as I can tell, so it seems unlikely to work - from https://github.com/SakanaAI/DiscoPOP/blob/main/scripts/launc... Unusually, the issue section doesn't exist so I can't provide feedback to them that way. But luchris429's repo does have it so will…

Author here! Thanks for pointing that out. The correct model name is indeed "gpt-4" instead of "gpt_model = 'gpt4_20231230_1106preview'". We were previously using an Azure endpoint, which is why the model name is different.

While I understand the frustration, I assure you that the rest of the code is functional. This was a simple oversight and should be a trivial fix. I appreciate your feedback and understanding.

Re: Can LLMs invent better ways to train LLMs?

#30

Post by s1gsegv »

seydor wrote: post #12

can LLMs optimize anything?

Absolutely. LLMs get a lot of hate but sincerely, GPT-4o can be given a hunk of code (one function/small class worth) and told to find optimization opportunities, and it will do a great job, especially considering the 30s it takes to ask,

It’s not perfect, but it understands lock-free algorithms, branch prediction, can tell you which memory order to use for atomic operations if you’re using too strong of an ordering, AND it will catch silly bugs at the same time. I had a bounds check in a lock-free algorithm I was optimizing, which equated to if(idx = end) return false, and it mentioned that error while optimizing.

This guy was really putting it through its paces on already highly optimized code in an esoteric architecture (Nintendo 64) and it still found a few things: https://youtu.be/20s9hWDx0Io

Post reply on HN