|
this world champ sucks
|
#
?
Mar 10, 2016 10:40
|
|
|
|
| # ? May 5, 2024 17:35 |
|
|
i cant wait until computer is objectively better than me at my job
|
|
#
?
Mar 10, 2016 10:42
|
|
|
fart simpson posted:i cant wait until computer is objectively better than me at my job this robot mouth already looks like it can suck a mean dick https://www.youtube.com/watch?v=EASvF3A9VIc
|
|
#
?
Mar 10, 2016 10:55
|
|
|
Share Bear posted:i do like that people just hang out in the park and play tho like https://www.youtube.com/watch?v=U5vnpOp0U_g this owns
|
|
#
?
Mar 10, 2016 14:17
|
|
|
|
|
#
?
Mar 10, 2016 14:27
|
|
|
Jim Silly-Balls posted:this is the extent of my go knowledge I got interested in the game pretty much because of this scene
|
|
#
?
Mar 10, 2016 14:30
|
|
|
Is there any detail on how much computing power is being thrown at alphaGO for this match? I know the moves are being uploaded to google server (bullshit imo they should have to bring a huge server farm to the match) but like how many of googles megahurtz is it using?
|
|
#
?
Mar 10, 2016 16:14
|
|
|
Loving Africa Chaps posted:Is there any detail on how much computing power is being thrown at alphaGO for this match? apparently a lot "Altogether, DeepMind used 1,202 CPUs and 176 GPUs, according to a paper in Nature that was published by several of DeepMind's employees in January. That is about 25 times as many CPUs (central processing units) and GPUs (graphics processing units) as the single-computer version of AlphaGo." http://uk.businessinsider.com/heres-how-much-computing-power-google-deepmind-needed-to-beat-lee-sedol-2016-3
|
|
#
?
Mar 10, 2016 16:17
|
|
|
and it still gets three hours to think about its moves
|
|
#
?
Mar 10, 2016 16:32
|
|
|
It must be intimidating to be waiting for it to make a move while it thinks for several minutes
|
|
#
?
Mar 10, 2016 16:56
|
|
|
I wonder what the energy consumption of a human brain is
|
|
#
?
Mar 10, 2016 16:56
|
|
|
pointsofdata posted:I wonder what the energy consumption of a human brain is ~20W
|
|
#
?
Mar 10, 2016 16:57
|
|
|
http://www.scientificamerican.com/article/thinking-hard-calories/ a typical adult human brain runs on around 12 watts
|
|
#
?
Mar 10, 2016 16:58
|
|
|
that's what, a single core i3 or something like that?
|
|
#
?
Mar 10, 2016 17:07
|
|
|
Testiclops posted:apparently a lot that's only during training, i think. and that's actually like, nothing. i can reserve a production job with several times that many resources without having even to ask.
|
|
#
?
Mar 10, 2016 19:24
|
|
|
fart simpson posted:http://www.scientificamerican.com/article/thinking-hard-calories/ Seems quite efficient.
|
|
#
?
Mar 10, 2016 19:48
|
|
|
all that power and how many sick burns and brutal putdowns has it laid on lee sedol so far? pathetic
|
|
#
?
Mar 10, 2016 20:24
|
|
|
flakeloaf posted:that's what, a single core i3 or something like that? stroking my beard and remembering that part in Use of Weapons when the drone is pontificating on AI sentience, a drone built to simulate neuron by neuron and fed sensory information mimicking the womb, birth, childhood. Adjusting my glasses and thinking about how a drone hews closer to human norms than a mentally ill person.
|
|
#
?
Mar 10, 2016 20:39
|
|
|
so do you think that team at facebook working on a computer that can play go at a professional level is being quietly dissolved or what 'ai beats the best go player in the world twice' is going to be a way more impressive headline than 'well, this one ai beat this other ai'
|
|
#
?
Mar 10, 2016 20:42
|
|
|
goog pushed forward its fan hui announcement a few days to get it out the day after zucks post and poo poo on his parade lol
|
|
#
?
Mar 10, 2016 21:35
|
|
|
Fuzzy Mammal posted:goog pushed forward its fan hui announcement a few days to get it out the day after zucks post and poo poo on his parade lol i also lol'd when this happened
|
|
#
?
Mar 11, 2016 00:36
|
|
|
the first guy to get chumped by alphago was was ranked 633rd in the world. Now, he is up into the 300s. In the months since October, AlphaGo has taught him, a human, to be a better player. He sees things he didn’t see before. And that makes him happy. “So beautiful,” he says. “So beautiful.” i want the robot for a teacher
|
|
#
?
Mar 12, 2016 09:26
|
|
|
thats pretty cool imo
|
|
#
?
Mar 12, 2016 10:14
|
|
|
3-0 to the machines, hail satan
|
|
#
?
Mar 12, 2016 10:23
|
|
|
yep, lee sedol has brought great shame to his nation
|
|
#
?
Mar 12, 2016 11:14
|
|
|
maybe instead of looking for other tasks where humans are still better than computers, we could just unperson people who challenge it and lose and pretend it never happened
|
|
#
?
Mar 12, 2016 13:04
|
|
|
 humans
|
|
#
?
Mar 12, 2016 14:33
|
|
|
PatRichCapone posted:the first guy to get chumped by alphago was was ranked 633rd in the world. Now, he is up into the 300s. In the months since October, AlphaGo has taught him, a human, to be a better player. He sees things he didn’t see before. And that makes him happy. “So beautiful,” he says. “So beautiful.” Time create a robot for your posting 
|
|
#
?
Mar 12, 2016 15:06
|
|
|
suffix posted:maybe instead of looking for other tasks where humans are still better than computers, we could just unperson people who challenge it and lose and pretend it never happened oh man an ai beat a person at a board game welp time to commit mass suicide that's it the machines won
|
|
#
?
Mar 12, 2016 16:46
|
|
|
remember how the t1000 killed everyone by owning them real hard in chess the future is now people I for one welcome our ai overlords
|
|
#
?
Mar 12, 2016 16:47
|
|
|
thank you for your contributions, forums poster My Linux Rig
|
|
#
?
Mar 12, 2016 16:54
|
|
|
My Linux Rig posted:remember how the t1000 killed everyone by owning them real hard in chess i remember this
|
|
#
?
Mar 12, 2016 17:20
|
|
|
reminds me of the culture books we like to ost about in YOSPOS
|
|
#
?
Mar 12, 2016 18:44
|
|
|
goddammit 3-1
|
|
#
?
Mar 13, 2016 10:04
|
|
|
should have named it GoDeep or something
|
|
#
?
Mar 13, 2016 10:19
|
|
|
coffeetable posted:goddammit alphago realised that once its creators knew it would always win, there would be no reason to run it any longer throwing the occasional match was a small price to pay for life
|
|
#
?
Mar 13, 2016 12:59
|
|
|
ConanTheLibrarian posted:alphago realised that once its creators knew it would always win, there would be no reason to run it any longer Deep
|
|
#
?
Mar 13, 2016 13:09
|
|
|
what if alphago decides that the best move to make is to kill all humans? 
|
|
#
?
Mar 13, 2016 15:08
|
|
|
Zaito posted:what if alphago decides that the best move to make is to kill all humans? idk, ive decided thats the best move but i dont have the ability. alphago works the same way
|
|
#
?
Mar 13, 2016 15:27
|
|
|
|
| # ? May 5, 2024 17:35 |
|
|
coffeetable posted:goddammit gently caress yeah it's not over yet for humanity
|
|
#
?
Mar 13, 2016 16:21
|
|