Jump to content

Welcome to FutureTimeline.forum
Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more. If you already have an account, login here - otherwise create an account for free today!

These ads will disappear if you register on the forum


  • Please log in to reply
167 replies to this topic

#1
Yuli Ban

Yuli Ban

    Nadsat Brat

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 17,143 posts
  • LocationAnur Margidda

Alright, I'm creating a dedicated thread for this group. Because now they're just flooding the AI & Robotics Thread. This is just getting ridiculous.
 
CLKUoAZ.jpg

http://www.reddit.co...ught_is_headed/

This is the company Google bought earlier this year. Quite frankly, it shows that things are moving faster with deep learning than they ever have with older models.


Edited by Yuli Ban, 05 August 2017 - 03:26 PM.

Nobody's gonna take my drone, I'm gonna fly miles far too high!
Nobody gonna beat my drone, it's gonna shoot into the sky!

#2
FutureGuy

FutureGuy

    Member

  • Members
  • PipPipPipPipPipPip
  • 557 posts

That's great. First, videogames, then, the world. Hopefully.



#3
tierbook

tierbook

    Member

  • Members
  • PipPipPipPipPipPipPip
  • 1,195 posts
  • LocationCharleston sc

Just imagine getting an army of these to be a games alpha testers. would find so many bugs if it really exploits everything.



#4
Ru1138

Ru1138

    Member

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 3,190 posts
  • LocationIllinois

http://www.technolog...turing-machine/

 

One of the great challenges of neuroscience is to understand the short-term working memory in the human brain. At the same time, computer scientists would dearly love to reproduce the same kind of memory in silico.

 

Today, Google’s secretive DeepMind start-up, which it bought for $400 million earlier this year, unveils a prototype computer that attempts to mimic some of the properties of the human brain’s short-term working memory. The new computer is a type of neural network that has been adapted to work with an external memory. The result is a computer that learns as it stores memories and can later retrieve them to perform logical tasks beyond those it has been trained to do.


What difference does it make?


#5
Unity

Unity

    Information Organism

  • Members
  • PipPipPipPipPipPipPipPip
  • 2,443 posts
Wow. That is some amazing work.

#6
star0

star0

    Member

  • Members
  • PipPipPipPipPip
  • 493 posts
Also see this posting from last week:

http://www.futuretim...will-look-like/

#7
Guest_Northward_*

Guest_Northward_*
  • Guests

The computer is currently being developed by the London-based DeepMind Technologies, an artificial intelligence firm that was acquired by Google earlier this year. Neural networks — which will enable the computer to invent programs for situations it has not seen before — will make up half of the computer’s architecture. Experts at the firm hope this will equip the machine with the means to create like a human, but still with the number-crunching power of a computer, New Scientist reports.
In two different tests, the NTM was asked to 1) learn to copy blocks of binary data and 2) learn to remember and sort lists of data. The results were compared with a more basic neural network, and it was found that the computer learned faster and produced longer blocks of data with fewer errors. Additionally, the computer’s methods were found to be very similar to the code a human programmer would’ve written to make the computer complete such a task.
These are extremely simple tasks for a computer to accomplish when being told to do so, but computers’ abilities to learn them on their own could mean a lot for the future of AI.
Elon Musk is not going to be happy about this.

 

 
Google's new computer programs itself



#8
SG-1

SG-1

    Member

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 3,653 posts
  • LocationUS - Arkansas
Holy sheet

Hey.  Stop reading.  The post is over.


#9
EVanimations

EVanimations

    Member

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 3,446 posts
  • LocationThe Eldritch Beyond

It happen



#10
star0

star0

    Member

  • Members
  • PipPipPipPipPip
  • 493 posts
More info here:

http://www.futuretim...will-look-like/
  • SG-1 likes this

#11
Yuli Ban

Yuli Ban

    Nadsat Brat

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 17,143 posts
  • LocationAnur Margidda
An amazing development, though without Very Short Term Memory it might still lapse compared to human memory.
Nobody's gonna take my drone, I'm gonna fly miles far too high!
Nobody gonna beat my drone, it's gonna shoot into the sky!

#12
Ru1138

Ru1138

    Member

  • Members
  • PipPipPipPipPipPipPipPipPip
  • 3,190 posts
  • LocationIllinois

Google Deep Mind is taking Artificial Intelligence to a new level and hope to accelerate scientific progress and truly useful robotics

Demis Hassabis leads what is now called Google DeepMind. It is still headquartered in London and still has “solve intelligence” as its mission statement. Roughly 75 people strong at the time it joined Google, Hassabis has said he aimed to hire around 50 more. Around 75 percent of the group works on fundamental research. The rest form an “applied research team” that looks for opportunities to apply DeepMind’s techniques to existing Google products.
Over the next five years, DeepMind’s technology could be used to refine YouTube’s recommendations or improve the company’s mobile voice search.

They dream of creating “AI scientists” that could do things like generate and test new hypotheses about disease in the lab. When prodded, he also says that DeepMind’s software could also be useful to robotics, an area in which Google has recently invested heavily.


What difference does it make?


#13
Yuli Ban

Yuli Ban

    Nadsat Brat

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 17,143 posts
  • LocationAnur Margidda
I know Star0 has something to add to this.
Nobody's gonna take my drone, I'm gonna fly miles far too high!
Nobody gonna beat my drone, it's gonna shoot into the sky!

#14
Zaphod

Zaphod

    Esteemed Member

  • Members
  • PipPipPipPipPipPip
  • 611 posts
  • LocationUK

I think that by the 2020's, Demis Hassabis and DeepMind will be known almost as much as Elon Musk and SpaceX. He is clearly a genius and is gathering other geniuses together. Google have the foresight to recognise this and essentially acquired this group of 'minds' for just £400 million.



#15
sasuke2490

sasuke2490

    veteran gamer

  • Members
  • PipPipPipPipPip
  • 464 posts

so why did they spend hundreds of millions for this guy what are possibilities could we get strong ai?


https://www.instagc.com/scorch9722

Use my link for free steam wallet codes if you want


#16
Zaphod

Zaphod

    Esteemed Member

  • Members
  • PipPipPipPipPipPip
  • 611 posts
  • LocationUK

so why did they spend hundreds of millions for this guy what are possibilities could we get strong ai?

 

Well other than the Human Brain Project they are one of the few companies that are researching and developing AI with a strong synthesis of both computation and neuroscience. This is key and is probably the fastest way in which we will be capable of producing self-improving intelligence.

 

Secondly, look at Demis Hassabis hugely impressive CV:

 

http://en.wikipedia..../Demis_Hassabis

 

Their website seems highly secretive http://deepmind.com/, so little is known of what they've actually come up with. Aside from this:

 



#17
Zaphod

Zaphod

    Esteemed Member

  • Members
  • PipPipPipPipPipPip
  • 611 posts
  • LocationUK

Vicarious is another company working on AI. It has recieved funding from Mark Zuckerberg and Jeff Bezos.



#18
star0

star0

    Member

  • Members
  • PipPipPipPipPip
  • 493 posts

Deepmind embedding intelligence into real-world robots; Hassabis theorizes creativity not so mysterious


Mr Hassabis confirmed that his team is starting to embed intelligence into real-world robots as well as developing tools to help humans understand the mysterious way the machines “think”.


I discussed the possibility in the comment thread here:

http://www.kurzweila...-five-years-off



#19
Yuli Ban

Yuli Ban

    Nadsat Brat

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 17,143 posts
  • LocationAnur Margidda

It can't be possible that people aren't losing their shit about this. I will make people lose their shit about this.
 
Artificial Intelligence Beats Go
Google Go-Playing Computer Beats Pro For First Time
Mastering the game of Go with deep neural networks and tree search
Google's AI Masters the Game of Go a Decade Earlier Than Expected
Google's DeepMind achieves AI 'breakthrough' by beating Go champion
 

 
Ars Technica's article writes that

Unlike previous computer game programs like Deep Blue, Deepmind doesn't use any special game-specific programming. For Breakout, an AI was given a general-purpose AI algorithm input from the screen and the score, and it learned how to play the game. Eventually Deepmind says the AI became better than any human player. The approach to AlphaGo was the same, with everything running on Google's Cloud Platform.

 
So this program wasn't written specifically for Go. It was just written and taught.

 
Here's a comment on the Futurology subreddit:
https://www.reddit.c..._top_go/czehtr1

As someone who has played Go for close to 25 years and a computer science degree from Yale, this shit is kind of scary.
 
Chess was solved by basically building a tree of all possible moves, something computers are really good at. Playing GO involves actual learning and recognition of complex patterns, in the number of which exceeds the number of atoms in the entire universe. I have always thought beating the best in GO will be a important mark in the progress towards true AI.

 
 
Another great quote:
http://www.nature.co...-defeat-1.19255  

The thing that struck me, playing through the games you couldn’t tell who was the human and who was the computer. With a lot of software you find the computer makes a lot of sensible moves and suddenly loses the plot. But here, you couldn’t tell which was which.

 

Just a couple of years ago, in fact, most Go players and game programmers believed the game was so complex that it would take several decades before computers might reach the standard of a human expert player


  • eacao, Casey and Whereas like this
Nobody's gonna take my drone, I'm gonna fly miles far too high!
Nobody gonna beat my drone, it's gonna shoot into the sky!

#20
TranscendingGod

TranscendingGod

    2020's the decade of our reckoning

  • Members
  • PipPipPipPipPipPipPip
  • 1,551 posts
  • LocationGeorgia

Yuli, people are not losing their wits because at first glance it does not seem like much. 2 factors play into this.

 

A lot of western people do not recognize the complexity of this game and a such do not understand the accomplishment that this is.

 

Secondly people will look back at Deep Blue winning at chess and IBM Watson winning at Jeopardy and say "oh, this is just another one of those little tricks"

 

You see people do not realize the magnitude of these accomplishments because they don't see the impact in their daily lives or even in something relatable.

 

Without a doubt this is momentous and it is one of those things that happens with exponential growth. This is a step and in a few more we will have machines with a level of intelligence we could never have imagined. 


  • eacao, Yuli Ban, ddmkm122 and 1 other like this

The growth of computation is doubly exponential growth. 






Also tagged with one or more of these keywords: DeepMind, deep learning, deep reinforcement learning, progressive neural network, artificial intelligence, AGI, differentiable neural, Google, RankBrain, artificial neural networks

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users