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

Rich Sutton on AI creativity and discovery

twitter.com

Post reply on HN
Print view View original

81–90 of 141 posts

Re: Rich Sutton on AI creativity and discovery

#81

Post by graphime »

jejeej wrote: post #76
artistonn wrote: post #67

Earlier quoted context omitted.

> art is ultimately about conveying human feelings and emotions you made a small error, art is mostly about generating an emotion in the viewer/listener/.... not about transmitting an emotion of the creator the Wikipedia page on art starts with: > Art is a diverse range of cultural activity centered around works utilizing creative or imaginative talents, which are expected to evoke a worthwhile experience https://en.…

No art is a means of expression - the creator uses an object as a vehicle to capture and distribute an expression. Also lol @ using Wikipedia as some source of absolute truth.

[dead]

Re: Rich Sutton on AI creativity and discovery

#82

Post by porridgeraisin »

singularity2001 wrote: post #73
musebox35 wrote: post #52

The most successful applications like coding are not the result of pure LLM/generative modeling. They come from closing the loop with an agentic harness. The generate-test-selectively refine loop is the core modality of scientific work. An LLM + RL with Verifiable Rewards + feedback from compiler/terminal runs mimics this process to a great extend. This is Fisher/Box feedback loop ( https://www-sop.inria.fr/members/I…

Most importantly, the reinforcement loop is used during training. I don't agree with Sutton's original hypothesis, but it holds even less after reinforcement learning.

RLVR still does not expand beyond the base distribution though, it only mode-seeks within it.

i.e, evaluation, retention yes. variation or "planning" no.

That is not to say you cannot use LLMs. Alpha evolve does exactly that. It uses an external simple evolutionary planner. The overarching point he's making is that our planner is still "dumb" and we need to work on it.

When you iteratively guide an LLM in claude code, you are the external planner. That also works.

Re: Rich Sutton on AI creativity and discovery

#83

Post by porridgeraisin »

highfrequency wrote: post #18

Unless I'm missing something, this argument seems to apply only to the original pretraining era (eg GPT 1-4). The post-training and reinforcement learning paradigms are clearly doing variation, evaluation and selective retention no?

RLVR still does not expand beyond the base distribution though, it only mode-seeks within it.

i.e, evaluation, retention yes. variation or "planning" no.

That is not to say you cannot use LLMs. Alpha evolve does exactly that. It uses an external simple evolutionary planner though. The overarching point he's making is that our planner is still "dumb" and we need to work on it.

When you iteratively guide an LLM in claude code, you are the external planner. That also works.

Re: Rich Sutton on AI creativity and discovery

#84

Post by magicalist »

musebox35 wrote: post #56
rembicilious wrote: post #14

"So that is my call to arms. If we want the full power of AI scientists, then we should share the goals with them so they can create, evaluate, discover, and in these ways fully participate in achieving the goals. Let’s be bold! Let’s fully automate Creativity and Discovery!" Should we automate exercise and play as well? How about learning? The machine didn't have a soul, so we donated ours. Eureka! My AI found it!

I understand the skepticism. I am worried about the implications of AI as well. The deeper issue at stake is that the depth of scientific knowledge has been increasing for a very long time. Now you get to have a PhD in esoteric subproblems and that slows down research especially if the discoveries require depth in multiple subdomains. Socially and economically training people in every combinatorial combination of sub…

Eh, it seems increasingly clear that no matter the fusion breakthroughs to come, it's going to be complicated and big and therefore expensive compared to renewables. By the time we figure it out it'll be a nice breakthrough but will end up on the shelf.

Re: Rich Sutton on AI creativity and discovery

#85

Post by artistonn »

jejeej wrote: post #76
artistonn wrote: post #67

Earlier quoted context omitted.

> art is ultimately about conveying human feelings and emotions you made a small error, art is mostly about generating an emotion in the viewer/listener/.... not about transmitting an emotion of the creator the Wikipedia page on art starts with: > Art is a diverse range of cultural activity centered around works utilizing creative or imaginative talents, which are expected to evoke a worthwhile experience https://en.…

No art is a means of expression - the creator uses an object as a vehicle to capture and distribute an expression. Also lol @ using Wikipedia as some source of absolute truth.

yeah, a bet a random HN-er definition of Art is much more reliable than Wikipedia

Re: Rich Sutton on AI creativity and discovery

#86

Post by ACCount37 »

TheOtherHobbes wrote: post #80
ACCount37 wrote: post #77

Earlier quoted context omitted.

Sounds like a skill issue? Recent image models are advancing rapidly at prompt adherence specifically, and being able to iterate on the same image is propelling them even further. Images 2.0 being the poster child of this "agentic iterative image composition" approach.

Images 2.0 isn't anywhere close to the kind of detail control I'm talking about. It's the opposite of a skill issue. No image generator is anywhere near the ballpark of pro-level manual Photoshop or Illustrator editing for individual elements in an image. If you don't understand this, try precisely kerning the text in a generated book cover to handle letter combinations like A and V. This is one of the big problems w…

I fail to see anything other than a skill issue.

We went from "AI can't generate text that isn't at least 20% typos and it always looks like shit" to "some letter combinations aren't kerned to perfection sometimes and adjusting that with prompts is hard". In a couple of generations.

Re: Rich Sutton on AI creativity and discovery

#87

Post by tines »

rembicilious wrote: post #14

"So that is my call to arms. If we want the full power of AI scientists, then we should share the goals with them so they can create, evaluate, discover, and in these ways fully participate in achieving the goals. Let’s be bold! Let’s fully automate Creativity and Discovery!" Should we automate exercise and play as well? How about learning? The machine didn't have a soul, so we donated ours. Eureka! My AI found it!

It seems human beings spend a significant portion of their lives trying to figure out how to live as little as possible.

Re: Rich Sutton on AI creativity and discovery

#88

Post by data_maan »

Another famous dude dumping his thoughts on HN who is gulping it up like an addict.

Add this to the long list of names like Terence Tao, and others who seem to be intellectually incontinent lately in the sense that one cannot navigate this space anymore without encountering their thoughts

Re: Rich Sutton on AI creativity and discovery

#89

Post by chadgpt3 »

musebox35 wrote: post #56
rembicilious wrote: post #14

"So that is my call to arms. If we want the full power of AI scientists, then we should share the goals with them so they can create, evaluate, discover, and in these ways fully participate in achieving the goals. Let’s be bold! Let’s fully automate Creativity and Discovery!" Should we automate exercise and play as well? How about learning? The machine didn't have a soul, so we donated ours. Eureka! My AI found it!

I understand the skepticism. I am worried about the implications of AI as well. The deeper issue at stake is that the depth of scientific knowledge has been increasing for a very long time. Now you get to have a PhD in esoteric subproblems and that slows down research especially if the discoveries require depth in multiple subdomains. Socially and economically training people in every combinatorial combination of sub…

Note that all fusion energy is unsustainable as it irreversibly turns water into helium.

Re: Rich Sutton on AI creativity and discovery

#90

Post by redblacktree »

chadgpt3 wrote: post #89
musebox35 wrote: post #56

Earlier quoted context omitted.

I understand the skepticism. I am worried about the implications of AI as well. The deeper issue at stake is that the depth of scientific knowledge has been increasing for a very long time. Now you get to have a PhD in esoteric subproblems and that slows down research especially if the discoveries require depth in multiple subdomains. Socially and economically training people in every combinatorial combination of sub…

Note that all fusion energy is unsustainable as it irreversibly turns water into helium.

We do have a LOT of water though. And surplus helium would be nice.
Post reply on HN