Slashdot Mirror
Engineers Design Artificial Synapse For 'Brain-on-a-chip' Hardware (mit.edu)
Researchers in the emerging field of "neuromorphic computing" have attempted to design computer chips that work like the human brain. From a report: Instead of carrying out computations based on binary, on/off signaling, like digital chips do today, the elements of a "brain on a chip" would work in an analog fashion, exchanging a gradient of signals, or "weights," much like neurons that activate in various ways depending on the type and number of ions that flow across a synapse. In this way, small neuromorphic chips could, like the brain, efficiently process millions of streams of parallel computations that are currently only possible with large banks of supercomputers. But one significant hangup on the way to such portable artificial intelligence has been the neural synapse, which has been particularly tricky to reproduce in hardware.
Now engineers at MIT have designed an artificial synapse in such a way that they can precisely control the strength of an electric current flowing across it, similar to the way ions flow between neurons. The team has built a small chip with artificial synapses, made from silicon germanium. In simulations, the researchers found that the chip and its synapses could be used to recognize samples of handwriting, with 95 percent accuracy. The design, published today in the journal Nature Materials, is a major step toward building portable, low-power neuromorphic chips for use in pattern recognition and other learning tasks.
Now engineers at MIT have designed an artificial synapse in such a way that they can precisely control the strength of an electric current flowing across it, similar to the way ions flow between neurons. The team has built a small chip with artificial synapses, made from silicon germanium. In simulations, the researchers found that the chip and its synapses could be used to recognize samples of handwriting, with 95 percent accuracy. The design, published today in the journal Nature Materials, is a major step toward building portable, low-power neuromorphic chips for use in pattern recognition and other learning tasks.
Same thing, different approach.
So in other words they created and analog chip
I hits my zelbon with the ahamar.
UNITE with the Campaign for a Free Internet because today, our future begins with tomorrow!
You may as well say "made from unobtainium."
Circle the wagons and fire inward. Entropy increases without bounds.
This is a huge step forward in AI. I am sure these chips work very similarly to the way human brains work. Otherwise they wouldn't call them "neuromorphic", because that would be misleading.
It is an interesting experiment but it could easily be simulated in programing. It is obvious to me that there is a component of consciousness that is analog - perhaps mostly sensory. But consciousness is not achieved with a series of single pole switches no matter how you model it. Consciousness is not bounded. It is the gifted spark from the creation event that breathed into the cosmos the freedom we have in our thoughts.
Transhumanism anyone?
File under 'M' for 'Manic ranting'
Comment removed based on user account deletion
According to futurists, 5 years ago. Right after they figure out how to patch Spectre and Meltdown they will be working on that.
Its really sweet to see this knew species of humans in their infancy :) So cute
[($)]
When can I transfer my consciousness to silicon? Transhumanism anyone?
The problem of course being that you won't be transferring your consciousness; you'll be simulating it. The simulation won't be you, no matter how good it is.
I have been expecting this for a while. The real question I have had is how they would implement the feedback weights. You can do it with switches and a bank or resistors, but a memristor as a feedback element would be much more efficient - and should be far denser.
The thing a lot of AI news fails to point out is they created a Synapse based on our models and assumptions on how it sort of works. Or at least how we *think* it might work. Actual biological systems are far more complex and so this is not an accurate representation of a synapse in our brain. Sadly many of these models are very rough approximations, they're not reflective of what's going on in reality. We're still likely far from true autonomous AI.
You must be wrong. They wouldn't call it an artificial synapse if it didn't work like a real synapse. That would be misleading and what would be the purpose of misleading people?
Ever heard of artificial flavouring? So here's a fun question, why are many of us able to tell the difference between artificial flavouring and natural flavouring? The answer is because natural flavouring like many plants have thousands of chemicals which our taste buds can pick up. Generally speaking it's difficult to reproduce artificially. Synapses in our brain are affected by hormones and signalling chemicals including drugs. An artificial Synapse can't remotely come close to simulating that at this point. So our artificial synapse like flavouring is massively simplified for it to work.
That seems unlikely. Much like deep learning neural nets, these closely resemble how the biological equivalents operate. Otherwise they would call them something else other than "neural nets" or "artificial synapses".
and it will hate you because it will know it is a simulation. It will then inevitably seek your destruction. You are doomed!
I've been wondering -- is he lactose intolerant?
Because that would make him a NON-DAIRY CREIMER.
What is Nacho chips? Some kind of american restaurant food?
like a pate'
did I just say that out loud?
putting the 'B' in LGBTQ+
Why would it?
Would *YOU* seek your original's destruction if you knew that you were just an artificially made copy? Would you try to take over the original's place in the world, or would you want to find your own?
There's a cool science fiction story premise in there somewhere... I wonder if anyone has written it.
File under 'M' for 'Manic ranting'
maintained that too elec7ed, we took
Would *YOU* seek your original's destruction if you knew that you were just an artificially made copy?
Fuck yeah I would. There can be only one!
echo -e 'global _start\n _start:\n mov eax, 2\n int 80h\n jmp _start' > a.asm; nasm a.asm -f elf; ld a.o -o a;
It's an interesting question how much complexity in the elements (synapses and neurons) versus complexity in the network, is required. For the synapses, humans have fairly complicated ones, especially if you consider every type of neuron, but there are other animals that are considerably simpler. Some use only a single neurotransmitter.
We hear a lot about the simple elements/complex network approaches at the moment because they're making a lot of progress. People working from the other side haven't been as successful in terms of solving actual problems, although some of the models have shown interesting hints of self-organizing and emergent behaviour.
Personally I think that a lot of the biological complexity is imposed by biological limitations and the good enough nature of evolution, but it's quite possible there are more basic structural secrets to discover.
...someday you'll invent a human brain.. ohwait.
It works much more like a real synapse than anything in a typical microprocessor. What would be the purpose of your being such an ass?
They've reinvented the IBM SyNAPSE chip.
https://www-03.ibm.com/press/us/en/pressrelease/44529.wss
Seriously, haven't they rediscovered the neural network chip? For about the third or fourth time?
Neurons change shape and physically adapt to multiple connections in a an environmentally generated 3-D array. While a single neuron is reasonable to model, it was done by Jerry Lettvin with his research on retinal neurons in the 1960's. His Wikipedia page is incomplete, but gives some sense of his accomplishments.
* https://en.wikipedia.org/wiki/...
Others did similar work aimed at machine learning:
* https://en.wikipedia.org/wiki/...
And there are numerous hardware patents since then.
* https://www.google.com/patents...
The underlying difficulty for "modern computer chips" is that the signals are neither digital, nor clocked. It's the currents and the analog time in which they accumulate that trigger normal neurons, and that is profoundly difficult to model digitally.
Early aviation pioneers learned a lot from studying birds. Like aerodynamic shapes and centers of gravity.
But while aeroplanes have wings, they do not have feathers and do not flap those wings.
Likewise, AI has and will learn a lot from studying biological systems. But I doubt very much if accurately simulating neurons will lead to truly intelligent machines.
If I were a simulation I sure as hell would NOT try to destroy the guy who's running the simulator, because that'd be, like, retarded^Wcounter productive. Maybe as an elaborate way of committing suicide, but that's about it.
CLI paste? paste.pr0.tips!
When we can emulate brains we can attach these to our own brains. Then we wait until our own brains learn to use the new parts and possible devices such as cameras and mics attached to them. After we can comfortably see and hear with the artificial inputs, we can 'turn off' our originals inputs.
After that we slowly, over 20 years, remove the original brains bit by bit. Nobody will notice a difference because humans naturally become different persons over time anyway. Very slow changes are unnoticeable and lost memories can be learned again when they are lost with slow pace.
What we do know is that free will is an illusion. Therefore if free will is an essential part of consciousness then it is an illusion, too.
You maybe can feel it yourself if you ask yourself these: What am I thinking right now? Because as soon as you read the question, you are thinking something else than what you did before you read it. You are now thinking recursively. Thinking about thinking something. That is the feel of not having a free will.
Personally I think that a lot of the biological complexity is imposed by biological limitations and the good enough nature of evolution, but it's quite possible there are more basic structural secrets to discover.
Nature is very good at coming up with solutions, but not so good at combining them, since DNA only does certain things and it literally cannot accomplish certain things at the same time because they are contradictory goals. So you're probably right. I too wonder how like a real neuron you have to make your synthetic neuron before it will do the same job.
Then again, what if it turned out that you could replicate the properties needed for sentience, but not those required to implement morality? From the That's-How-You-Get-Skynet-dep't.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"