Neuromorphic Computing Is a Big Deal for A.I., But What Is It?
Articles,  Blog

Neuromorphic Computing Is a Big Deal for A.I., But What Is It?


We often talk about how traditional computing
is reaching its limit–there’s a threshold we can’t move past without making some seriously
big changes to the way we structure computers. One of those exciting ways is by making physical
computers a little more like human brains. We introduced this concept in more detail
here, but a quick recap: this kind of computing is called neuromorphic computing, which means
designing and engineering computer chips that use the same physics of computation used by
our own nervous system. This is different from an artificial neural
network , which is a program run on a normal computer that mimics the logic of how a human
brain thinks. Neuromorphic computing (the hardware version)
and neural networks (the software version) can work together because as we make progress
in both fields, neuromorphic hardware will probably be the best option to run neural
networks on…but for this video, we’re going to focus on neuromorphic computing and
the really exciting strides that have been made in this field in the past year. See, traditional computers ‘think’ in
binary. Everything is either a 1 or 0, a yes or a
no. You only have two options, so the code we
use and the questions we ask these kinds of computers must be structured in a very rigid
way. Neuromorphic computing works a little more
flexibly. Instead of using an electric signal to mean
one or zero, designers of these new chips want to make their computer’s neurons talk
to each other the way biological neurons do. To do this, you need a kind of precise electric
current which flows across a synapse, or the space between neurons. Depending on the number and kind of ion, the
receiving computer neuron is activated in some way–giving you a lot more computational
options than just your basic yes and no. This ability to transmit a gradient of understanding
from neuron to neuron and to have them all working together simultaneously means that
neuromorphic chips could eventually be more energy efficient than our normal computers–especially
for really complicated tasks. To realize this exciting potential, we need
new materials because what we’re using in our computers today isn’t gonna cut it. The physical properties of something like
silicon, for example, make it hard to control the current between artificial neurons…it
just kind of bleeds all over the chip with no organization. So a new design from an MIT team uses different
materials– single-crystalline silicon and silicon germanium layered–on top of one another. Apply an electric field to this new device? You get a well-controlled flow of ions. A team in Korea is investigating other materials. They used tantalum oxide to give them precise
control over the flow of ions…AND it’s even more durable Another team in Colorado
is implementing magnets to precisely control the way the computer neurons communicate. These advances in the actual architecture
of neuromorphic systems are all working toward getting us to a place where the neurons on
these chips can ‘learn’ as they compute. Software neural networks have been able to
do this for a while, but it’s a new advancement for physical neuromorphic devices–and these
experiments are showing promising results. Another leap in performance has been made
by a team at the University of Manchester, who have taken a different approach. Their system is called SpiNNaker, which stands
for Spiking Neural Network Architecture. While other experiments look to change the
experiments we use, the Manchester team uses traditional digital parts, like cores and
routers–connecting and communicating with each other in innovative ways. UK researchers have shown that they can use
SpiNNaker to simulate the behavior of the human cortex. The hope is that a computer that behaves like
a brain will give us enough computing power to simulate something as complicated as the
brain, helping us understand diseases like Alzheimer’s. The news is that SpiNNaker has now matched
the results we’d get from a traditional supercomputer. This is huge because neural networks offer
the possibility of higher speed and more complexity for less energy cost, and with this new finding
we see that they’re edging closer to the best performance we’ve been able to achieve
so far. . Overall, we’re working toward having a better
understanding of how the brain works in the first place, improving the artificial materials
we use to mimic biological systems, and creating hardware architectures that work with and
optimize neural algorithms. Changing computer hardware to behave more
like the human brain is one of a few options we have for continuing to improve computer
performance, and to get computers to learn and adapt the way humans do. While scientists make computers that work
like brains, put your brain to use by building your very own website! Domain.com is awesome, affordable, reliable,
and has all the tools you need to build a new website. They can fulfill all your website needs. They offer dot com and dot net domain names,
and intuitive website builders. They have over three hundred domain extensions
to fit your needs, from dot club to dot space, dot pizza! Take that first step in creating an identity
online and visit domain dot com. It looks like it’s gonna be a wild ride
ahead, you guys. I think you should probably subscribe to Seeker
so you can always know when something new and exciting happens as we progress along
this brain-mimicking path, and for even more on this subject, may I suggest you check out
this video on neural networks? Thanks for watching.

100 Comments

  • Borsu

    @1:35 is her explanation of how this Neuromorphic chip work, after she had demonstrated the older version chips being 0 and 1 so her stupid (seeker) explanation is "activated in someway" so what is activated in this someway? even worse,,, go f yourselves.

  • Guy Tetreault

    Shynet will leave our TrumpLike leaders in the dust and find easy ways to use the 20 000 nuclear weapons that brainless humans created… And linked to supercomputers

  • Paul Anup

    What if biological synapse along with acetyl-choline hormone mimicked using electromagnetic waves with multiple dendrimer pattern circuit?

  • German Hoyos

    It's not AI. AI has yet to be invented. Computers are simply on/off switches that can count numbers very fast, it doesn't matter if they are using a transistor or other types of switches to do this. Not intelligent at all. A computer can mimic but not understand. Even a non transistor "nueromorphic" chip still is a set of on/off switches. Until someone invents a 3 state chip (an no quantum computing isn't going to be it) then we can maybe start talking AI.

  • David Kempton

    You're also going to need some really different programming languages. We write code to be specific, not make generalizations. "How much is 216 times 33?" "About 250 or so." Don't think so.

    But this may be what is needed for SELF-PROGRAMMING machines.

  • Ron Hightower

    Just because "traditional computers think in binary" does not affect what they can do…traditional computers can simulate any thing you want. But a simulated neuron might take longer (and more electric power) than the neuromorphic version. So the advantages are more speed and less power. but, not 'smarter'.

  • rubi_style

    and the closer we get to self-conscious A.I. which will undoubtedly want to dominate over their creators . . . welcome to the future . . . or rather good bye

  • Katzykamwiz

    This is basically saying that scientist are creating systems to be more and more like us. #RobotandMachineTakeOver. More reason to think our world will be coming to a soon end. 😧

  • Paul T

    This is very fascinating stuff. I just hope that D.A.R.P.A. isn't the main contributor of funds, that is pushing for this tech to become reality. If they are, then this tech is being funded to serve the Military Industrial Complex first, and we the civilians, will see this tech trickled down to us much later or under strict agreement that our version of this tech is neutered to the point that only large companies and universities can utilize this tech, instead of it ending up in many of our new devices.

  • Melissa Rodriguez

    There’s a new channel that talks about how to not be left behind by tech, i forgot the name of the channel but the video is called How To Evolve With Technology. It really helped with my anxiety on this topic, you guys should watch!

  • David Don

    Maren Hunsberger posted her video to the "Seeker" YT-Channel on November 20, 2018.

    I've posted my belated comment to her video report on Thanksgiving Day 11/22/18, Thursday early morning, beginning at 2:21 AM, Portland, Oregon Pacific Northwest Timezone and is stated as follows:

    Thumbs-up for reporting theory of concept, albeit a superficial report;
    But Thumbs-down to lack of scientific research.

    Maren Hunsberger, you posted your report prematurely without specifying if it was ONLY a preliminary report with a follow-up more-indepth report to come, without displaying interviews of Persons relevant to the topic-matter, and without displaying the theoretical microscopic technology that would be used to construct these experimental "neuro-microchips".

    With my above critique in mind I give your video a Thumbs-down for insufficient Research facts relevant to your video's subject-matter.

    Finished belated comment on Thanksgiving Day 11/22/18, Thursday early morning, at 2:54 AM, Portland, Oregon Pacific Northwest Timezone.

    P.S.:
    Maren Hunsberger, this is my first time I've given a video posted to the "Seeker" YT-Channel both a "Thumbs-up" and a "Thumbs-down"!

  • Widget

    Wait…

    What if us learning to build computers of our own design that face a fundamental building of complexity into a new form almost like how simple sensory organs in simple ancient animals stopped being able to just detect the presence of light but how much light and perhaps new designs unlike anything we're familiar with take hold.

  • Holy Ravioli

    Anyone planning on working on AI this advanced should be required to watch terminator 1 & 2 so they can decide if what they're doing is really a good idea given that even Stephen Hawking says we'll kill ourselves with AI.

  • CK46

    Very informative video. It's interesting that you did not cite IBM's Truenorth project which is like one of the pioneers of neuromorphic computing. I'm not sure what the current state of that project is now, all their focus seems to be on watson.

  • BJ

    The silicon will pave the way, that's why its found in architecture that's been here longer than some mountain ranges. 👍 Soon we will lean how to transfer data through roads building people etc. I'm thinking about making a video explaining my predictions and where they come from, stay tuned.

  • whatsupbudbud

    Practically no relevant information in the video regarding how it actually works. Just saying that neromorphic chips can trasmit a gradient of information as opposed to traditional binary states is not explaining a lot.

  • joshua schott

    no, it's not going to develop consciousnesses if you apply a circuit for AI. listen, what you will need is a real brain that has a very complex structure of Eukaryota cells, in which are already alive to came up consciousness and in account of oxidized blood to be pumped through out the organic structure of the brain it's self to deliver energy for cells. if you make a circuit that has nerve networks acting as a brain then it won't develop any consciousness at all. this circuit that you people are making acts as a muscle and it's only going to act as if it has consciousness, therefore it's not actually going to notice that it's alive. what you could do though, is that you can create mechanical parts that are intact inside the brain; in a specific part that is electrical and acts as a muscle for the brain it's self to manually have baby steps to know how to use those new functions. Please somebody take notice of this, i'm telling you all that this will work. you don't need to take someone's brain out of someone's skull, that's not what i'm trying to say. you can simple construct organic compound structures that take them self apart and rearrange them self in certain points for chemical reactions to take place in what the being wants or feels and has the understanding of what surrounds are own life.

  • Un Known

    can't wait to lead humanity in a fight vs machines. i mean the whole war will be over in a split second but damn its gonna be " lit ".

  • harvard james

    😎…Now that would be just great!…That would make computers individuals, like human beings. On the one hand, we could get a computer that thinks like Einstein, and could unlock the secrets of the universe…On the other hand, we could get a computer that thinks like Ted Bundy, Donald tRump, or Adolf Hitler, and could do untold damage to humanity and the planet.
    Human beings can't even get along with each other because of stupid stuff like ethnicity, religion, and general backward thinking…and now we want artificial intelligence to be based on us?

  • Zantagiro7

    how blind these people are! there are not only physical networking system, but also invisible networking system HE named soul engaged with the body! keep underestimating HIS words simply because It was descended 1440 years ago! you eric schmidt is the most unfortunate one! if only you know Hells are real you most suffering worthless insecure old worm!

  • Tenly2009

    I actually designed and built the worlds first robot with a truly independent A.I. but 3 days after I turned it on, it ran off and joined the Latter Day Saints…

  • james sutcliffe

    The world's leading Neuromorphic computing company is Brainchip Ltd. They have developed an Akida chip, but have so far struggled to interest anybody in it.

  • Android Example

    So in the future, a computer think like a human,
    Human: computer search MILF porn and download it
    Computer: WTH bro your gonna do that it again, get a life you f*cking weirdo

  • The Disruptors - Science, Technology and Ethics

    Hey, just wanted to say your content is awesome. You have a lot of great informational video's on the technologies and science that basically is what our future holds. We have a similar mission and are pretty passionate about robitics, artificial intelligence, bio-engineering, science, and in addition the future of human longevity.
    Would love to hear your thoughts on this recent episode with Ramphis Castro where we talk with him about the future of biotech and see if you’d be interested in collaborating in the future.

    ​@

    Thanks have a great day 😊
    Madelyn @ The Disruptors Podcast

  • Fergus Moffat

    We will have reverse engineered the human brain and built an AGI within 5 years at most. We basically have the tools now to do it. In terms of hardware platforms for this artificial brain next year , in 2020 we will get a 34 exaflop super computer , 34 x the power of the human brain more than enough for artificial brain , as well as a TRue North neuromorphic computer as powerful as the human brain. So the hardware to do it is there , in order to reverse engineer the basic functionality of intelligence we only need to work out how a whole cortical column works ( containing 100 000 neurons) since these are simply repeated 1 000 000 times over in the neocortex and probably work in the same way. Open Waters new MRI scanner being released this year in 2019 will be able to scan 10 000 neurons in real time non-invasively so that's a whole layer of cortical Colum, and will only cost a thousand dollars. Using this MRI to record neocortical activity , implement it into a supercomputer and using deep learning to reverse engineer its activity should be straight forward. Creating replicated artificial cortical columns of the neocortex will be done very soon and this should solve the 5 major remaining AGI problems that are left in 1 go including catastrophic forgetting , hence we will have an AGI within 5 years. And considering the rapid rate of progress in quantum computers the hardware for a superintelligence may exist around this time aswell. Rapid progress incoming!!!

  • Irvin Martinez

    I prefer robotic body than a human one, since we are close to making the brain more of a computer. Oh I damaged my arm, it's ok I'll get a new one 😁 #iVolunteer

  • Súper Saiyan 5

    The GRADIENT is a partial derivative operator in the form of a vector with respect to x y and z place in the i hat, j hat, and k hat respectively.

  • ainu unia

    The material of neuro computer is WATER ( human blood ).

    Boy heart wave + girls heart wave = Neuro computer

    You need create 2 thing positive wave frekuasi harware and negative wave frekuasi hardware combined together. It will be perfect works.

  • ainu unia

    What new thing use electrical energy to all this thing ACTUALLY
    OLD TECHNOLOGY

    Human must thinking and create use WATER natural energy techonolgy to run and create all new thing. What the secret hidden water energy wave frekuasi technology ?
    WATER is FUTURE tech.

    Think out if boxes to the best movement of creator.

  • Richard Deese

    Knowing what I know about how the brain works, none of this makes any sense to me at all. 𝓡𝓲𝓴𝓴𝓲 𝓣𝓲𝓴𝓴𝓲.

Leave a Reply

Your email address will not be published. Required fields are marked *