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

Pac-Man recreated with a GAN trained on 50k game episodes

blogs.nvidia.com

Post reply on HN
Print view View original

81–90 of 154 posts

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#81

Post by lumost »

lostmsu wrote: post #76
lumost wrote: post #70

I've observed an increasing trend with modern AI research that A) It's no longer practical to incorporate into real products given the hardware costs for training and inference. In NLP, large transformer based models can easily occupy a 5 thousand dollar GPU at a paltry 50-60 words per second processing pace. B) The Research often demonstrates things that existing non-ML toolchains were already reasonable at doing wi…

A) is just a sign, that further engineering, both model and framework, is required. Also, 50-60 words per second does not sound too bad. It amounts to 4M words per day. I imagine most people only type a thousand words per day. So with a single GPU you can handle 4000 users (assuming batch-style processing). B) This specific example actually shows ML toolchains are orders of magnitude better, than existing ones. Imagi…

On A: Using a p3.2xl for that task leaves you with a minimum ec2 cost of goods sold of 50 cents per user per month for the target task. To make a reasonable SaaS business out of this you would need to be charging a minimum of $5.00 per user month for the service that this GPU is providing, assuming all other tasks have a negligible impact on Cost of Goods Sold, and that you are able to spread the load optimally throughout the day - and thus don't have to purchase higher capacity during peak.

This isn't too bad of a situation if this GPU provides 100% of your value proposition. but, if you needed your product to perform 5 such tasks - or competition/customer demand forces you to move to a bigger model with a 10 words per second inference rate then you may find it hard to balance customer value vs. price vs. margin.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#82

Post by baq »

simias wrote: post #51
itchyjunk wrote: post #8

Thought experiment: Is my brain also generating reality without a physics engine and is filling in details that it thinks should be there? If true, what are the advantages/disadvantages of it?

The other answers to your comment are great but I also want to point out that you can feel your brain doing that if you play some videogames for a long time. You can feel that you develop an intuition for the way the game will behave even if the in-game physics are very different from our real word, showing the plasticity of our brain. If we had a fully hardcoded physics simulator in our head we couldn't "think with…

I felt exactly this when playing Portal for the first time when it came out. It felt like nothing before or after. Highly recommended.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#83

Post by YayamiOmate »

YeGoblynQueenne wrote: post #77
lumost wrote: post #70

I've observed an increasing trend with modern AI research that A) It's no longer practical to incorporate into real products given the hardware costs for training and inference. In NLP, large transformer based models can easily occupy a 5 thousand dollar GPU at a paltry 50-60 words per second processing pace. B) The Research often demonstrates things that existing non-ML toolchains were already reasonable at doing wi…

Or, to paraphrase: great paper, no application.

Like every so-called basic science. 1st batteries, radio, lasers etc also looked like that.

I see a strong parallel. It's a poor performance but important proof of concept.

Also, it's general in computing trend since assembly. Why compile when you can write machine code, waste of resources. Why have abstract compiler, when you can write assembly. Why have higher level language when it cant generate code optimal for target when it doesnt know it. Then web came, it's not faster than 20 years ago because with increased computing power bloat came. Somehow it always turns that seemingly wasteful apporach outweights its cons. Empirically lowered entry cost and increased applicability has very very high value. At least higher that I used to assess.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#84

Post by esoterica »

lumost wrote: post #70

I've observed an increasing trend with modern AI research that A) It's no longer practical to incorporate into real products given the hardware costs for training and inference. In NLP, large transformer based models can easily occupy a 5 thousand dollar GPU at a paltry 50-60 words per second processing pace. B) The Research often demonstrates things that existing non-ML toolchains were already reasonable at doing wi…

Do you really think nVidia is seriously advocating for using neural networks as the best way to write a PacMan clone? It's obviously just a proof of concept to demonstrate that GANs are sophisticated enough to learn certain categories of models.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#85

Post by empath75 »

lumost wrote: post #70

I've observed an increasing trend with modern AI research that A) It's no longer practical to incorporate into real products given the hardware costs for training and inference. In NLP, large transformer based models can easily occupy a 5 thousand dollar GPU at a paltry 50-60 words per second processing pace. B) The Research often demonstrates things that existing non-ML toolchains were already reasonable at doing wi…

The ai has created a functional internal model of an external system based on limited sensory input. That is essentially a kind of consciousness.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#86

Post by YeGoblynQueenne »

SQueeeeeL wrote: post #80
YeGoblynQueenne wrote: post #77

Earlier quoted context omitted.

Or, to paraphrase: great paper, no application.

It really is a great paper, the problem is when your entire field has few real applications but everyone is making too much money to admit that. Cool tech demos like this is awesome as a blog post, but not sure why it's academic...

Well, the paper has an experiment where they train an agent based on their system to play two of the three games (PacMan, PacManMaze and VizDoom) and compare its result to two other systems (an LSTM-based one and a model-based RL one, WorldModel). Their system gets the best score in Pacman but unfortunately the version that does better than WorldModel is the one that doesn't have the memory module whose usefulness is a central claim of their paper. WorldModel outperforms their system in VizDoom by ~300 points, though their system without the memory module "solves" VizDoom (i.e. scores just a bit over than 750; that is 765) and they claim that it's the first Gan-based system that achieves this.

A bit thin of the ground (especially when it comes to supporting the claim about the usefulness of the memory module) but the way neural nets research and especially RL research is going (from what I've seen, not my area of expertise) that's the kind of result that you can expect will get your paper published these days.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#87

Post by esoterica »

YeGoblynQueenne wrote: post #77
lumost wrote: post #70

I've observed an increasing trend with modern AI research that A) It's no longer practical to incorporate into real products given the hardware costs for training and inference. In NLP, large transformer based models can easily occupy a 5 thousand dollar GPU at a paltry 50-60 words per second processing pace. B) The Research often demonstrates things that existing non-ML toolchains were already reasonable at doing wi…

Or, to paraphrase: great paper, no application.

There is a reason why "Research" and "Development" are listed as separate concepts when people talk about R&D. The point of most research papers is not to directly invent a working product, it's to produce fundamental insights that may be useful in future applications.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#88

Post by eckza »

empath75 wrote: post #85
lumost wrote: post #70

I've observed an increasing trend with modern AI research that A) It's no longer practical to incorporate into real products given the hardware costs for training and inference. In NLP, large transformer based models can easily occupy a 5 thousand dollar GPU at a paltry 50-60 words per second processing pace. B) The Research often demonstrates things that existing non-ML toolchains were already reasonable at doing wi…

The ai has created a functional internal model of an external system based on limited sensory input. That is essentially a kind of consciousness.

> The ai has created a functional internal model of an external system based on limited sensory input.

Sure.

> That is essentially a kind of consciousness.

That is a flying leap.

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#89

Post by SkyBelow »

lumost wrote: post #70

I've observed an increasing trend with modern AI research that A) It's no longer practical to incorporate into real products given the hardware costs for training and inference. In NLP, large transformer based models can easily occupy a 5 thousand dollar GPU at a paltry 50-60 words per second processing pace. B) The Research often demonstrates things that existing non-ML toolchains were already reasonable at doing wi…

It reminds me of older electric cars. They worked. They could get you around. But they did so at higher price, slower speeds, or other factors that a consumer wasn't willing to pay for. Thus they were regulated to novelty items that some work campuses would have and concept cars that would be shown from time to time.

But eventually we reached the point where they are not only competitive, they may be positioned to fully overtake combustion cars.

I wonder how many discoveries on electric cars ended up being fully dead ends along the path that took us to the point we are now?

Re: Pac-Man recreated with a GAN trained on 50k game episodes

#90

Post by esoterica »

SQueeeeeL wrote: post #80
YeGoblynQueenne wrote: post #77

Earlier quoted context omitted.

Or, to paraphrase: great paper, no application.

It really is a great paper, the problem is when your entire field has few real applications but everyone is making too much money to admit that. Cool tech demos like this is awesome as a blog post, but not sure why it's academic...

I don't know where you got the idea that the purpose of an academic paper is to produce the schematics for a working commercial product. The fact that it's novel and interesting is sufficient grounds for producing a paper about it. Why do you think projects like these are solely the domain of blog posts?
Post reply on HN