|
https://deepmind.com/blog/wavenet-generative-model-raw-audio/
|
# ? Sep 12, 2016 22:39 |
|
|
# ? May 27, 2024 02:04 |
|
can people talk about neural networks. poo poo owns. im way to dumb to get it
|
# ? Sep 12, 2016 22:40 |
|
update your e/n thread mate
|
# ? Sep 12, 2016 23:24 |
|
echinopsis posted:can people talk about neural networks. poo poo owns. im way to dumb to get it neural networks are a way to make computers think
|
# ? Sep 13, 2016 03:25 |
|
fart simpson posted:neural networks are a way to make computers think a..learning computer?
|
# ? Sep 13, 2016 03:26 |
|
I just wanna be able to gently caress around with these things
|
# ? Sep 13, 2016 03:58 |
|
echinopsis posted:I just wanna be able to gently caress around with these things http://playground.tensorflow.org/#a...&hideText=false
|
# ? Sep 13, 2016 05:17 |
|
echi update ur e/n thread plz. also there is a fyad thread about it now.
|
# ? Sep 13, 2016 11:11 |
|
jeeze
|
# ? Sep 13, 2016 11:23 |
|
my sex life has generated huge levels of interest forum wide. but this thread is for neural networks
|
# ? Sep 13, 2016 11:29 |
|
single responsibility principle
|
# ? Sep 13, 2016 11:55 |
|
i try to avoid thinking about stuff tbh
|
# ? Sep 13, 2016 12:26 |
your mom who is also my teacher and I did some deep learning
|
|
# ? Sep 13, 2016 15:25 |
but realtalk neural networks own and they can play videogames https://www.youtube.com/watch?v=qv6UVOQ0F44
|
|
# ? Sep 13, 2016 15:26 |
|
It's just minimizing error functions using straightforward curve-fitting math. The high level concepts were published in CS papers decades ago and basically outline strategies for applying gradient descent a bunch of times really fast in various combinations. The difference now is that computers are faster and we have user-friendly libraries and better input data. The internal states of the trained model AKA the 'thinking' parts are meaningless gibberish, just tables of coefficients for reducing data in highly specific input contexts, and just happen to work better than most people expected for certain types of problems but don't fundamentally bring us closer to true AI, OP.
|
# ? Sep 13, 2016 16:29 |
|
echinopsis posted:my sex life has generated huge levels of interest forum wide. but this thread is for neural networks what if we had a neutral network that was for pleasuring your wife?
|
# ? Sep 13, 2016 18:42 |
|
I wasn't hoping for true Ai. just cool gimmicks like the thing where it takes a paining and applies the style to goatse etc
|
# ? Sep 13, 2016 20:02 |
|
echinopsis posted:I just wanna be able to gently caress around
|
# ? Sep 13, 2016 20:38 |
|
echinopsis posted:can people talk about neural networks. poo poo owns. im way to dumb to get it you're too dumb to notice things under your nose
|
# ? Sep 13, 2016 20:42 |
|
same
|
# ? Sep 13, 2016 20:42 |
|
echinopsis posted:my sex life has generated huge levels of interest forum wide. but this thread is for neural networks update your e/n thread then
|
# ? Sep 13, 2016 21:46 |
|
oh wait, you did! god bless
|
# ? Sep 13, 2016 21:46 |
|
https://www.youtube.com/watch?v=zP5J6a14CR0
|
# ? Sep 13, 2016 21:52 |
|
OSI bean dip posted:you're too dumb to notice things under your nose mate was this called for
|
# ? Sep 13, 2016 22:21 |
|
this one time i thought about a man FOR A WHILE
|
# ? Sep 14, 2016 01:41 |
|
Buttmeister posted:but realtalk neural networks own and they can play videogames video games are just low-res simulated realities
|
# ? Sep 14, 2016 01:53 |
|
duTrieux. posted:video games are just low-res simulated realities you sound drunker than i am
|
# ? Sep 14, 2016 01:57 |
|
emoji posted:It's just minimizing error functions using straightforward curve-fitting math. The high level concepts were published in CS papers decades ago and basically outline strategies for applying gradient descent a bunch of times really fast in various combinations. The difference now is that computers are faster and we have user-friendly libraries and better input data. The internal states of the trained model AKA the 'thinking' parts are meaningless gibberish, just tables of coefficients for reducing data in highly specific input contexts, and just happen to work better than most people expected for certain types of problems but don't fundamentally bring us closer to true AI, OP.
|
# ? Sep 14, 2016 07:26 |
|
Buttmeister posted:but realtalk neural networks own and they can play videogames training neural networks with genetic algorithms? your optimizer is a piece of poo poo
|
# ? Sep 14, 2016 07:37 |
sorry my flux capacitor is in the shop LOL
|
|
# ? Sep 14, 2016 13:43 |
|
mate, these fuckers just can't get past you loving rather than answer the more important questions itt
|
# ? Sep 17, 2016 07:32 |
|
gently caress
|
# ? Sep 17, 2016 07:43 |
|
life is gay
|
# ? Sep 18, 2016 04:22 |
|
echinopsis posted:my sex life has generated huge levels of interest forum wide. but this thread is for neural networks
|
# ? Sep 18, 2016 07:07 |
|
echinopsis posted:mate was this called for echi i think i just created a fyad invasion of ur e/n thread since i posted a meltdown and full assault w/r/t the assholes trolls i may have hosed it up
|
# ? Sep 18, 2016 11:39 |
|
just lol if u think, op
|
# ? Sep 18, 2016 15:25 |
|
thinking isn't my strong suit, op
|
# ? Sep 18, 2016 15:39 |
|
computers take numbers in and spit numbers out. they are no smarter than a pencil. like a pencil, they can be manipulated to produce state that can be interpreted as intelligence, but the intelligence lies in the operator, not in the tool.
|
# ? Sep 18, 2016 18:11 |
|
but what if it grows on its own?
|
# ? Sep 18, 2016 18:23 |
|
|
# ? May 27, 2024 02:04 |
|
PALE SPECTRES posted:life is gay
|
# ? Sep 18, 2016 19:18 |