The mission of Google’s DeepMind Technologies startup is to “solve intelligence.” Now, researchers there have developed an artificial intelligence system that can mimic some of the brain’s memory skills and even program like a human.
The researchers developed a kind of neural network that can use external memory, allowing it to learn and perform tasks based on stored data. The so-called Neural Turing Machine (NTM) that DeepMind researchers have been working on combines a neural network controller with a memory bank, giving it the ability to learn to store and retrieve information.
The system’s name refers to computer pioneer Alan Turing’s formulation of computers as machines having working memory for storage and retrieval of data.
The researchers put the NTM through a series of tests including tasks such as copying and sorting blocks of data. Compared to a conventional neural net, the NTM was able to learn faster and copy longer data sequences with fewer errors. They found that its approach to the problem was comparable to that of a human programmer working in a low-level programming language.
Additional Coverage: http://phys.org/news/2014-10-google-deepmind-acquisition-neural-turing.html
Related: Neural Turing Machines http://arxiv.org/abs/1410.5401 and Learning to Execute http://arxiv.org/abs/1410.4615.
Related Stories
Researchers from Google subsidiary DeepMind have published an article in Nature detailing AlphaGo, a Go-playing program that achieved a 99.8% win rate (494 of 495 games) against other Go algorithms, and has also defeated European Go champion Fan Hui 5-to-0. The researchers claim that defeating a human professional in full-sized Go was a feat expected to be achieved "at least a decade away" (other statements suggest 5-10 years). The Register details the complexity of the problem:
Go presents a particularly difficult scenario for computers, as the possible number of moves in a given match (opening at around 2.08 x 10170 and decreasing with successive moves) is so large as to be practically impossible to compute and analyze in a reasonable amount of time.
While previous efforts have shown machines capable of breaking down a Go board and playing competitively, the programs were only able to compete with humans of a moderate skill level and well short of the top meat-based players. To get around this, the DeepMind team said it combined a Monte Carlo Tree Search method with neural network and machine learning techniques to develop a system capable of analyzing the board and learning from top players to better predict and select moves. The result, the researchers said, is a system that can select the best move to make against a human player relying not just on computational muscle, but with patterns learned and selected from a neural network.
"During the match against [European Champion] Fan Hui, AlphaGo evaluated thousands of times fewer positions than Deep Blue did in its chess match against Kasparov; compensating by selecting those positions more intelligently, using the policy network, and evaluating them more precisely, using the value network – an approach that is perhaps closer to how humans play," the researchers said. "Furthermore, while Deep Blue relied on a handcrafted evaluation function, the neural networks of AlphaGo are trained directly from gameplay purely through general-purpose supervised and reinforcement methods."
The AlphaGo program can win against other algorithms even after giving itself a four-move handicap. AlphaGo will play five matches against the top human player Lee Sedol in March.
Google and Facebook teams have been engaged in a rivalry to produce an effective human champion-level Go algorithm/system in recent years. Facebook's CEO Mark Zuckerberg hailed his company's AI Research progress a day before the Google DeepMind announcement, and an arXiv paper from Facebook researchers was updated to reflect their algorithm's third-place win... in a monthly bot tournament.
Mastering the game of Go with deep neural networks and tree search (DOI: 10.1038/nature16961)
Previously: Google's DeepMind AI Project Mimics Human Memory and Programming Skills
(Score: 1) by Megahard on Thursday October 30 2014, @10:14PM
How does it do on Jeopardy?
(Score: 2, Interesting) by Anonymous Coward on Thursday October 30 2014, @10:15PM
Could this technology be used to explain how the fuck some people can possibly think that systemd is a good idea? Could it give us some insight into the deviancy, and perhaps mental retardation, that such people may be suffering from?
(Score: 2) by wonkey_monkey on Thursday October 30 2014, @10:45PM
Maybe it would explain why people keep bringing up systemd in entirely unrelated discussions. Go and tell someone who gives a monkey's.
systemd is Roko's Basilisk
(Score: 0) by Anonymous Coward on Thursday October 30 2014, @10:51PM
Replace "systemd" with "incest" or "beastiality", if you want. The GP's comment is legit and totally on topic. One of the major benefits of the software described in the summary is its ability to help us understand the human mind and human thought processes. If it can be used to understand why some people with broken thought processes think the way they do, then it could be immensely beneficial to all of humanity.
(Score: 0) by Anonymous Coward on Thursday October 30 2014, @11:43PM
Replace "systemd" with "incest" or "beastiality", if you want. The GP's comment is legit and totally on topic. One of the major benefits of the software described in the summary is its ability to help us understand the human mind and human thought processes.
Wow, that's actually a very cogent response. Hats off to you, sir/madam/other!
P.S. The problem of rationalizing system D is NP-complete, so I wouldn't expect much, even from this type of advanced neural network system.
(Score: 2) by wonkey_monkey on Friday October 31 2014, @10:23AM
Replace "systemd" with "incest" or "beastiality", if you want.
I hope those aren't name suggestions for a new init system.
Anyway, the commenter clearly has no interest in the actual subject at hand. He's just shoe-horning in systemd because it's the in-thing to gripe about.
c.f. http://soylentnews.org/comments.pl?sid=4632&cid=111710 [soylentnews.org]
It's like Family Guy's Brian when he tries to turn a conversation around to something he desperately wants to share his opinion on.
"Oh, what was that? Did you just say 'self-help book'?"
"No."
"Because I just wrote one..."
systemd is Roko's Basilisk
(Score: 2, Funny) by Anonymous Coward on Friday October 31 2014, @12:02PM
incest = initialization control and execution system
beastiality = boot event and state transision initialization automation for Linux type systems.
(Score: 0) by Anonymous Coward on Friday October 31 2014, @12:06PM
I don't see how that comment you linked to is off topic. The submission is about attacks against software and computer systems. Malware could be used to perform such attacks. Systemd, due to its many, many flaws, is considered to be malware by many people. So that comment is totally on topic and relevant when it brings up systemd. Whoever voted it down incorrectly should never be allowed to moderate ever again.
(Score: 0) by Anonymous Coward on Friday October 31 2014, @04:11AM
For that to work you will need a CPU x3 more powerful than the ones currently on the market.... ;)
(Score: 0) by Anonymous Coward on Friday October 31 2014, @09:43AM
So you mean you just need three current CPUs to solve this problem? Sounds very affordable.
(Score: 2) by gringer on Friday October 31 2014, @05:15AM
See here [nih.gov]
Ask me about Sequencing DNA in front of Linus Torvalds [youtube.com]
(Score: 0) by Anonymous Coward on Friday October 31 2014, @11:49AM
Talk about the pot calling the kettle black!
(Score: 0) by Anonymous Coward on Friday October 31 2014, @12:04PM
OK: Could it be used to explain the pot calling the kettle black?
(Score: 0) by Anonymous Coward on Friday October 31 2014, @12:10PM
Please refrain from using racial slurs here.
(Score: 3, Interesting) by buswolley on Thursday October 30 2014, @10:43PM
"Mimics the brain"
Yawn.
Dont get me wrong. These kinds of models are cool, but answering "Why is this Novel?" requires more.
I'm a fan of hippocampal computational models for event memory
subicular junctures
(Score: 2) by Subsentient on Saturday November 01 2014, @01:06AM
We are the Borg. Lower your shields and surrender your ships. We will add your biological and technological distinctiveness to our own. Your culture will adapt to service us. You will be assimilated. Resistance is futile.
"It is no measure of health to be well adjusted to a profoundly sick society." -Jiddu Krishnamurti