Two Minute Papers
The paper “Neural Task Programming:
Learning to Generalize Across Hierarchical Tasks” is available here:
https://stanfordvl.github.io/ntp/
We would like to thank our generous Patreon supporters who make Two Minute Papers possible:
Andrew Melnychuk, Brian Gilman, Christoph Jadanowski, Dave Rushton-Smith, Dennis Abts, Eric Haddad, Esa Turkulainen, Evan Breznyik, Kaben Gabriel Nanlohy, Malek Cellier, Michael Albrecht, Michael Jensen, Michael Orenstein, Robin Graham, Steef, Steve Messina, Sunil Kim, Torsten Reil.
https://www.patreon.com/TwoMinutePapers
Two Minute Papers Merch:
US: http://twominutepapers.com/
EU/Worldwide: https://shop.spreadshirt.net/TwoMinutePapers/
Music: Antarctica by Audionautix is licensed under a Creative Commons Attribution license (https://creativecommons.org/licenses/by/4.0/)
Artist: http://audionautix.com/
Thumbnail background image credit: https://pixabay.com/photo-2137333/
Splash screen/thumbnail design: Felícia Fehér – http://felicia.hu
Károly Zsolnai-Fehér’s links:
Facebook: https://www.facebook.com/TwoMinutePapers/
Twitter: https://twitter.com/karoly_zsolnai
Web: https://cg.tuwien.ac.at/~zsolnai/
Source
Congratulations !! For 100,000 subs! In advance keep growing
A step towards "the final solution" >_> uh oh ….
Would be cool to see a video on capsule networks by Hinton
I think you should make the images more closely follow with what you are explaining at that moment. It sometimes feels like there are two separate streams of information because the video shows a different concept to the voice-over explanation.
Combine this , neural evolution and the theory of mind , and you have super human ai
Why are the robot movements so slow?
To my understanding, if the physical motor movement engineering is solved then actions' duration would be near immediate? I'd think that in a sim, without slowing down, it would take practically no time from begin to end state?
I know time is not currently the main goal of the research.
"It may be a valuable puzzle piece to build towards a final solution."
FINAL SOLUTION TO THE LAST PROBLEM OF HUMANITY
Nah we goodd, singularity is far away, no need to worry …watch video … well there goes human, bow to your future overlord
Sounds like recursion to me
Thanks for the uploads
Love your videos so much and I am so thankful that you keep them so frequent! Really thank you!
Robots will play with dominoes till the end of life.
Humans might just be the playing pieces.
Why does the robot arm keep going back? So wasteful.
Yesterday you mention a robot performing a backflip and today Boston Dynamics releases a video of just that.
Nem is értem, hogy miért nem találtam meg korábban ezt az oldalt! Remek tartalom, csak így tovább Károly! 😀
Tbh I feel the pursuit of general AI is exceptionally risky. Fair enough, some research obsessed humans are happy to be displaced in terms of capability completely by AI, but most people wont be. I think there needs to be serious research into ethics, or even investigations into encoding empathy or morality.