DeepMind’s New AI Dreams Up Videos on Many Topics

📝 The paper "Efficient Video Generation on Complex Datasets" is available here:

🙏 We would like to thank our generous Patreon supporters who make Two Minute Papers possible:
313V, Alex Haro, Andrew Melnychuk, Angelos Evripiotis, Anthony Vdovitchenko, Brian Gilman, Bruno Brito, Bryan Learn, Christian Ahlin, Christoph Jadanowski, Claudio Fernandes, Daniel Hasegan, Dennis Abts, Eric Haddad, Eric Martel, Evan Breznyik, Geronimo Moralez, James Watt, Javier Bustamante, John De Witt, Kaiesh Vohra, Kasia Hayden, Kjartan Olason, Levente Szabo, Lorin Atzberger, Lukas Biewald, Marcin Dukaczewski, Marten Rauschenberg, Maurits van Mastrigt, Michael Albrecht, Michael Jensen, Nader Shakerin, Owen Campbell-Moore, Owen Skarpness, Raul Araújo da Silva, Rob Rowe, Robin Graham, Ryan Monsurate, Shawn Azman, Steef, Steve Messina, Sunil Kim, Taras Bobrovytsky, Thomas Krcmar, Torsten Reil, Zach Boldyga.

Károly Zsolnai-Fehér's links:

[RobertMilesAI]

So cool. Really interesting failure modes too. Like in the top left at 2:06, it gets the falling down and bouncing up of the basketballs, but doesn't understand that each ball has to fall down before it can bounce up again.

[zeldamax]

Imagine AI understood story writing and cinematography. There could always be another episode on netflix, because it would be generated as soon as you click next.

If you were to feed the AI personal data aswell, it would specifically manufacture content for your preferences. Maybe this will happen in the next few decades?

[Kram]

Imagine, like, a transformer or, as here, a GAN (or some fancy combination of those), which you can hand a movie script and it'll put out a plausible movie with all the fancy shots and fully synthesized audio and everything you might want

[SadieTheMooingKitty]

Can't wait until we can generate endless po....video with this technology.

[scarm_rune]

were 1 step away from creating unlimited minecraft letsplays

[Reneator]

You could fill an art gallery with these videos. Because they challenge how you perceive reality. "The way an AI sees the world"

[cala]

would those generated videos be available, too? :) it's a pity there are only thumbnails

[sharathkumar]

Looks like the day isn't too far where we'll be enjoying synthetic programs on our virtual reality head sets. A la Fahrenheit 451.

[CappaBeta]

Did the AI imagine something obscene @2:10 ?? :'D Third from the right, four down. I have no idea what I'm looking at. It's very interesting reglardless!

[TracksWithDax]

MAN, this is so creepy! The generates imagery is a combination of being extremely familiar, things where you could actually describe what it’s meant to be, yet unrecognizable if you start to take more than just and glance at it. Amazing technology, but the stuff it creates creeps me out!

[ThatBidsh]

do you think it's possible that everything neural networks (and humans) do can be described either as pattern recognition or pattern generation? and that possibly we just need a sufficiently generalized solution for each (and improved hardware) to have human level or beyond general AI?

[eelcohoogendoorn]

2:23, top-right... seems like it is overfitting on electroBOOM outtros :P

[Sinebeast]

The possibilities of fake image in a couple papers down the line is fucking frightening.

[dunkus]

My brain hurts when I try to understand what I'm staring at

[Jadinandrews]

I am waiting for someone to use a GAN to generate software. I wonder if that's the first step towards a true strong AI; giving it the tools to improve its own code base?

[CapnSlipp]

Why would it take more papers to create a higher-res video? Just run the existing algorithm with a higher output layer and some tweaked NN settings, and let the computer run longer (or use more powerful hardware). If the algorithm can already scale from 64x64 to 256x256, it can scale to 2048x2048 (again, with some retraining/tuning for detail density). This is not a scientific problem, just an engineering task.

[EmileAI]

This is really worrisome. It's really scary

[neoqueto]

GANs are a dead end for AI research I think, they only creates output that looks convincing, but isn't "something", with proper context.

[raytsh]

The videos are so creepy, so uncanny, it nearly makes me insane.

[tamimyousefi]

Generated not sequentially? That's weird.