Комментарии:
I'm subbing this dude is amazing
ОтветитьThank you very informative❤❤
ОтветитьWhat would network lay out look like?
Ответитьi want to program this immediately, this is a great sign.
Ответитьyoure the best for making this i love you man
Ответитьyour intro is beautiful btw
ОтветитьOMG WTF HOW DID YOU KNOW MY FAVOURITE SONG?!!! AHHHHHH
ОтветитьVery intelligent, as someone who finds this very relevant, the rgb lights add on for pico 2 looks a good way to visualize this a little affordably. I am going to read on the 2016 paper. Geoffrey Hinton is amazing to hear beforehand. Thank you for creating this
ОтветитьThis is a wildly fantastic video
ОтветитьImagine your favorite song is viva la vida by Coldplay, and you DON’T want to kill yourself. Crazy, right?
ОтветитьNeed networking protocol to explain pls
ОтветитьNot to be pedantic, but systems’ tendency to minimize their potential energy is not directly related to the 2nd law of thermodynamics. The 2nd law states that closed systems increase in entropy over time; it does not deal with energy values directly.
Ответитьone of the most underrated channels.
ОтветитьTrying to understand this content, my brain hurts
ОтветитьThat’s really interesting theme, as a programmer I’m gonna try to create it by myself)
ОтветитьHi great video! Does someone know if a network like that would work based on qubit-like 'neurons' ? Would the connection with the weights lead to quantum decoherence? Would the storage capacity be greater because of the superposition of states?
ОтветитьIs it pronounced sin apse. Not sigh naps
ОтветитьSuper intuitive! Very well done! I will wait for the video on Modern Hopfield Network :P
ОтветитьHop Hop 😅
ОтветитьThanks !
ОтветитьI found this quite inspiring, thanks. Over the last few days it got me writing a few classes and methods in python to deal with Hopfield networks, so I could have a play myself. I've just used them to implement an 8x8 net for simple character training and recognition from the python command line. Though I know the concept has its limits, I found it enlightening to play with.
I've dabbled with simple layered networks a little bit years ago, with some success, but it was a bit hit and miss, and felt a bit "by magic". Though for a different structure, the video gave me some new insight into weight adjustment.
👍
(edit: Should be called "Associative Memory Networks" - see @tfburns post in this thread)
Дружище, связка отлично работает. Всем советую. Спасибо!
ОтветитьThis feels like it has overlap with Conway's game of life in terms of emergent behavior from a simple set of rules.
Ответитьnice video! thank you for your work
ОтветитьNobel prize winning video
ОтветитьMr Kirsanov, have you heard about the least action principle Euler-Lagrange equation and the path integral? Given that I'd not be too confident to say that a ball does not search or follow all possible trajectories before appearing to follow a parabola.
Ответитьwatching this video after the announcement of Physics Nobel 2024 :)
ОтветитьIt sounds like a really useful way to reduce the problem space. Well done!!
ОтветитьNobel Prize idea
ОтветитьThis video aged well :)
ОтветитьSaw this video a month ago and even watched it twice. Just had to come back and drop a comment now that the legend’s got a Nobel under his belt.
ОтветитьBeautifully explained. Thank you
Ответитьi am here again after the Nobel .
doing research in this domain for living but this video is always refreshing .
High quality and useful, thank you.
ОтветитьI feel this is a little misleading, like never heard of dictionaries? self balancing trees? hashing?
Ответитьgreat video , love the way of thinking thing through
Ответитьwow, very insightful, very brilliant.
Ответить支
ОтветитьNobel prize brought me here
Ответить