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

Genie: Generative Interactive Environments

sites.google.com

Post reply on HN
Print view View original

11–17 of 17 posts

Re: Genie: Generative Interactive Environments

#11

Post by nullptr_deref »

jparkerholder wrote: post #7
polygamous_bat wrote: post #5

Earlier quoted context omitted.

What is the video resolution, 64x64? And even then it becomes blurry. Seems like another Google flag-plant-y paper filled with hot air that we will never see the source code or model for because it will expose how poor its capabilities are relative to competitors. The internal politics at these places must be exhausting. Industry research was supposed to be free from the publish or perish mindset, but it seems like i…

Hey author here :) First, tough crowd, love it, always great to get feedback because we are actively working on improving the model. We are very happy to admit it is not perfect, but given not many people thought this was possible a year ago, I am quite excited to see the next step of improvement. This is like the GPT1 of foundation world models, and we have a fair few ideas in the works to speed up progress. The res…

How can I get started with this kind of research? Is it even possible without a PhD? Thanks.

Re: Genie: Generative Interactive Environments

#12

Post by jparkerholder »

polygamous_bat wrote: post #8
jparkerholder wrote: post #6

Earlier quoted context omitted.

Author here :) Re: 1) typically no, but of course it can hallucinate just like LLMs. 2) Agreed but the key point missing is Dreamer is trained from an RL environment with action labels. Genie is trained exclusively from videos and learns an action space. This is the first version of something that is now possible and will only improve with scale.

Thanks for braving the crowd here, you will unfortunately only find hard questions. Anyway, about my second question: why are the videos only half second ish long? Does the model unravel after that? Also > This is the first version of something that is now possible and will only improve with scale. 11b params is already pretty large considering the stable diffusion and LLM scale. How much higher do we need to scale u…

The bigger issue is lack of generating novel content rather than a total "unravel". We focus on OOD images because our motivation is generating diverse environments, but these are much harder to play for longer vs images closer to the training videos. It is interesting because one of the things you gain when going from 1B->10B is the OOD images working at all. Note it is not even trivial to detect the character given our model does not train with any labels or have any inductive biases to do so.

Point of clarification -- we don't expect bigger models to be the only way to improve this and are working on innovations on the modeling side, however we don't want to overlook the significance of scaling either :)

Re: Genie: Generative Interactive Environments

#13

Post by jparkerholder »

nullptr_deref wrote: post #11
jparkerholder wrote: post #7

Earlier quoted context omitted.

Hey author here :) First, tough crowd, love it, always great to get feedback because we are actively working on improving the model. We are very happy to admit it is not perfect, but given not many people thought this was possible a year ago, I am quite excited to see the next step of improvement. This is like the GPT1 of foundation world models, and we have a fair few ideas in the works to speed up progress. The res…

How can I get started with this kind of research? Is it even possible without a PhD? Thanks.

If we did a good job then the paper should be written in a way that is digestible. When you don't understand things, follow the references to learn more (and there's probably videos covering most of the components we use).

In the Appendix we have a case study that should be possible to re-implement and run with a single GPU/TPU. We are hoping the community can build from that and innovate. If you take these steps and get stuck, feel free to get in touch!

Re: Genie: Generative Interactive Environments

#14

Post by jparkerholder »

jasonjmcghee wrote: post #10

> Genie is capable of converting a variety of different prompts into interactive, playable environments that can be easily created, stepped into, and explored If these are generating a fully interactive environments, why are all the clips ~1 second long? Based on the first sentence in your paper, I would have expected a playable example as a demo. Or 20. But reading a bit further into the paper, it sounds like the mo…

That is correct! The model generates each frame on the fly.

Re: Genie: Generative Interactive Environments

#17

Post by YeGoblynQueenne »

jparkerholder wrote: post #12
polygamous_bat wrote: post #8

Earlier quoted context omitted.

Thanks for braving the crowd here, you will unfortunately only find hard questions. Anyway, about my second question: why are the videos only half second ish long? Does the model unravel after that? Also > This is the first version of something that is now possible and will only improve with scale. 11b params is already pretty large considering the stable diffusion and LLM scale. How much higher do we need to scale u…

The bigger issue is lack of generating novel content rather than a total "unravel". We focus on OOD images because our motivation is generating diverse environments, but these are much harder to play for longer vs images closer to the training videos. It is interesting because one of the things you gain when going from 1B->10B is the OOD images working at all . Note it is not even trivial to detect the character give…

>> Note it is not even trivial to detect the character given our model does not train with any labels or have any inductive biases to do so.

Why not add inductive biases then and make your life easier? What's with this choice to try and do everything the hard way, presumably to make a point? In the end the point made is so specific that it translates to nothing that is usable in real problems.

See MuZero for example- sure, you can learn without being given the rules explicitly, just from the win/loss signal, but then that only works in board games and atari games, and without the chance of a snowball in hell that it will work in the real world. We're dazzled by the technical prowess, but real utility? Where is that?

Post reply on HN