Caltech creates first artificial neural network from DNA

One of the things that our brains excel at is the ability to recognize what things are, even when presented with an incomplete set of data. If we know only that an animal is sold in pet stores and stuffs food in its cheeks, for instance, we can be pretty certain that the animal in question is a hamster. Now, for the first time ever, researchers at the California Institute of Technology (Caltech) have created a DNA-based artificial neural network that can do the same thing ... albeit on a very basic level. They believe that it could have huge implications for the development of true artificial intelligence.

Link

As a committed technophile one of the things I'm eagerly awaiting is an artificial AI capable of managing the planet and all of our resources as a whole. Such an intelligence removed of political and financial motivation would surly be better for us that depending on flawed humans with personal agendas to produce an abundance paradigm for us.

Recently Caltech has taken a massive step forward in this process and may have uncovered at least the fundamental building blocks of how to design intelligent self aware systems. This coupled with machines that surpass the cognitive levels of humans will truly usher in a new age of intellectual development and freedom for mankind.

I hope you enjoy.

reply to post by Imhotepsol


This also ties in well with recent advances where machines were able to learn language and skills by reading the manual (RTFM!). Two machines were able to learn how to play civilizations and install software.

Researchers at MIT's Computer Science and Artificial Intelligence Lab have been able to create computers that learn language by doing something that many people consider a last resort when tackling an unfamiliar task - reading the manual (or RTBM). Beginning with virtually no prior knowledge, one machine-learning system was able to infer the meanings of words by reviewing instructions posted on Microsoft's website detailing how to install a piece of software on a Windows PC, while another was able to learn how to play Sid Meier's empire-building Civilization II strategy computer game by reading the gameplay manual.

Link

I remember watching and episode of "Through The Wormhole" and there was a part where some scientists fused Neural tissue to a motherboard, linking it up to a computer, so the Neural network would "learn" from the actions given to it by the computer through the fused motherboard. Interesting s***

reply to post by mr10k


Yeah that was the living bot made from Rat neurons. I thought that was pretty freaky when I first saw it, later when I realized what they actually achieved I was just in awe of it. And this technology will adhere to moores law so every 18 months the power and efficiency should double.

Its only a matter of time until we have a global consciousness network we all interact with, that learns from us as we do from it. It'll be really interesting to see what way this technology goes. There's so much potential for good and evil to come from it but either way it'll be damn interesting.

Crazy stuff. Looks like mankind is getting ready to open all sorts of cans of worms in the near future. Not sure I like that.

I think I read a journal article about this in a class a year or two ago.

It's not all that impressive. DNA has been known to be able to infer from only partial information for years. That's how DNA repairs itself, and transcribes itself into RNA and eventually into proteins. If it couldn't read partial information, in fact, then there wouldn't be any control over transcription beyond an "on-off" toggle... when the starting point for transcription is only partially "correct", there is a lower (but nonzero) likelihood of successful transcription, thus decreasing transcription rate to a given fraction of its maximum.

In layperson's terms, DNA has been known to "infer" for years. This isn't a huge invention or discovery, and it's a very long ways from being able to infer about things more interesting than G, C, T, and A.

reply to post by pondrthis


You're absolute right but now they're coming up with methods to train the DNA if you will which will allow us to take advantage of DNAs potential in computational terms - which is something we previously were unable to do very well.

reply to post by Balkan


Could be scary. I like to think of it like this.

Imagine twenty years from now and you've lost someone close to you. Now imagine being able to enter a virtual reality where they are still there, as conscious and alive as they were here. Imagine we were able to indefinitely preserve our greatest geniuses and they could still learn and interact with the world long after the confines of a physical body would have given out.

As I said there is the potential for so much good to come from this but there's also the potential for the bad. I'd like to think that as we approach the singularity that we'll be able to utilize these technologies for the advancement of everyone. It will be pointless having class divisions, social barricades and so on once we get to the singularity.

reply to post by Imhotepsol


S&F
This is fanastic news. The advancements of the human race really do come in leaps an bounds. I guess skynet is just around the corner

It reminds me of the experiment done with the monkey with electrodes in his brain at Duke University that could make a pair of mechanized legs walk on a treadmill in Japan. If we could couple the two. That would make for very exciting times. Now the question is could we program selfaware neural networks to perform specific tasks with out them getting bord with their initial programming. For example; Multi-tooled farming equipment that can manage fields by themselves. And could we program them with the three laws of cybernetics? I could talk about this all day. Thanks for the post.

www.youtube.com...

reply to post by XLR8R


We're going to end up emulating God consciousness eventually. A truly omnipotent and omnipresent intelligence that has nothing but our advancement and best interests at 'heart' (as much as that's possible with a machine). Once we have that consciousness we could connect every other piece of technology we have to it as inputs so that it can take weather data from satellites and use that to plan agriculture, see if droughts will occurs, how to best manage the supply chain. Its not just limited to agriculture but can extend to every facet of human life and desire.

I'm like you mate this is getting very exciting because we're beginning to see all of these disparate pieces that have been put together over the last few years are coming together into new technologies that have the potential to change everything forever. There's no going back from this - failing a complete and utter global catastrophe in the next 5 years.

I dunno about these experiments, scary stuff.

I'm not sure I would want an AI running things for us either.

I'm kind of the opinion that living a life informs us and our decisions based on mistakes and learning from them.

A created thingy might still be concious and smart but would it have compassion? I think not.

The ant mentality is great to get stuff done, but things get left behind.

peace.

reply to post by treespeaker


You seem to forget we live in a world where we are ruled by psychopaths, fueled by god complexes and without compassion. Personally I'd rather a machine intelligence running things then flawed humans looking to aggrandize the self. That's just my opinion though.

reply to post by Imhotepsol


I couldn`t agree more. In the last 20 years so many manufacturing plants have become automated that we could patch those in as well. I used to work in the automotive industry and everything from motors to chasis were made almost entirely by robots. they would feed a molded aluminium block into a machine and it would bore all the holes for the cylinders and every other little thing you can imagine being fitted to the motor. Same goes for the painting, the safetyglass and many other tidbits. All done with algorythms. I don`t know if you ever heard of the venus project but I had looked at the inovations needed to bring this vision to reality and it seemed impossible with curent tech. But now this opens so many new doors, it might even make This ''Utopia`` possible after all. Lets keep imagination alive and see what dreams may come.

reply to post by XLR8R


Exactly. Everything would become an input to improve the system. It's like additional senses if you will. I think though that this change isn't going to happen slowly, its more probable that it will happen so quickly and we'll adapt to it so fast that the Utopian Society will look back and wonder WTF happened, how did we get from there to here.

Consider this. The average amount of information we generate per year is roughly equivalent to all of the advancement from the time of Christ to the renaissance, and that's also increasing exponentially every year. Our evolution now in one human lifetime is probably equal to 10 generations previously. We're evolving faster and faster and we're being wired to adapt to change faster than ever before, mostly due to access to information and technology. Once we finally merge with it then we will literally uncap the limits of development and our only limit will be our greatly enhanced imaginations.

Also I'm a big fan of the Venus project.

reply to post by treespeaker


I agree with you but when the first cave man started his own fire, I'm sure that was scary stuff as well. But look what direction that has taken us. New discoveries and evolution IS very scary. When man fist walk on the moon...to tell you the truth I don't if I could of gone through with it if I was in Neil Armstong's shoes. I could bet my life that Nikola Tesla had a few scary moments as well. Again, look at the advancements that gave us. We shoulden`t fear change, we should embrace it. And through trial and error we will provail. One thing we have to remember, humanity is still in it`s enfantsy and right now I think we`re going through the ``Terrible Two`s`` Very selfish and lack empthy. If we can get through these time times then humanitie`s future will be bright indeed.

reply to post by XLR8R


My favorite story about Tesla is this. Its just the deadpan way he delivered the line that killed me.

A reporter for the New York Herald wrote on March 13 that he came across the inventor in a small café, looking shaken after being hit by 3.5 million volts, "I am afraid," said Tesla, "that you won't find me a pleasant companion tonight. The fact is I was almost killed today. The spark jumped three feet through the air and struck me here on the right shoulder. If my assistant had not turned off the current instantly in might have been the end of me." Tesla, on contact with the resonating electromagnetic charge, found himself outside his time-frame reference. He reported that he could see the immediate past - present and future, all at once. But he was paralyzed within the electromagnetic field, unable to help himself. His assistant, by turning off the current, released Tesla before any permanent damage was done.

While many may or may not remember the DARPA Minds Eye project that was announced in January of 2011, that program was going to be contracted to various universities to develop the architecture and the programming needed to achieve and give surveillance cameras the ability to think and make decisions.

While that function may sound all military and needed for DOD security operations, but what many miss is that this Minds Eye program was to develop artificial intelligence and develop the means to incorporate such artificial intelligence into key functions that humans would do slower and with human hesitation due to moral or religious factors.

Such a Mind's Eye program is an effort to give the current weapon systems and machines the ability to think for themselves and to do what humans would hesitate to do.

This is why computer programming advances are needed to mesh or bring together artificial intelligence with the appropriate machines to achieve some technological advance that places America in the super power seat once again with technology that can not only do what humans would hesitate to do, but think while it does it.

Such a leap in technological advancement has many military,corporate and business uses, but it also has the potential for great evil.

This recent Cal-tech press release involving the first neural network from DNA is in part because Cal-tech was one of the 12 groups that were selected to work on the Mind's Eye Project. As such this latest announcement is a new announcement of how far Cal-tech has come since January of this year when such a program was begun as announced by DARPA on January 4, 2011.

Those that want to better understand the Mind's Eye program as it relates to this recent Cal-tech announcement should visit the following link to DARPA where the program is explained in greater detail and where it shows Cal-tech as being one of the major universities where such AI research was going to be conducted as part of the DARPA Mind's Eye project.

In fact, the link will show or name all the major military industrial complex corporations involved and the universities that such research is being developed at. It is worth a read in my opinion.

DARPA Mind's Eye link, helps to better explain the Mind's Eye project and why this latest Cal-Tech press release is an advancement in the DARPA Mind's Eye project by a university selected to conduct such DARPA research and development. Just thought you should know. Thanks again.

www.darpa.mil...

If you guys enjoy stuff like this then you should probably check out.

singularityhub.com...

They have lots of news articles about exactly this type of stuff..

reply to post by Imhotepsol


If what Tesla claimed was true it would explin quite a bit. It's a real shame that his efforts were scuttled do to greed and emperialism. If we would of went the Tesla way we would live on a much greener planet today and be far more advanced technologicly. The Venus project would Probably be our way of life by now, it being the next logical stepping stop for the advancements of our race. That said, since he came out of this unharmed, I would love to see pics of Tesla all messed do to this. It would be de hilarious since he mostly looks prestine in is pics.

What I wonder, if AI is running the show, what if AI figures out Man is a parasite to the Earth and the only way for the Earth to survive is to eliminate the parasite? Or as some elitists and upper government believe, the only way mankind can survive is a 95% percent reduction in humanity. I don't personally subscribe to this opinion in the whole. IMHO we are a parasite on the earth, but we are necessary to the Earth and other living entities to complete the chain.
Although that brings up a perspective I have debated before. If man were eliminated from the chain would the other life on Earth continue to thrive? Or would it be catastrophic like if the honey bees, or other similar examples, were removed from the chain?
I believe advanced AI development and neural network cybernetics is innevitable in the very near future.That is,of course, if the right people survive, without much technological degredation, the coming Eath changes.

Peace and harrnony
Don