Computer chips inspired by human neurons can do more with less power
Kwabena Boahen got his first computer in 1982, when he was a teenager living in Accra. “It was a really cool device,” he recalls. He just had to connect up a cassette player for storage and a television set for a monitor, and he could start writing programs.
But Boahen wasn’t so impressed when he found out how the guts of his computer worked. “I learned how the central processing unit is constantly shuffling data back and forth. And I thought to myself, ‘Man! It really has to work like crazy!’” He instinctively felt that computers needed a little more ‘Africa’ in their design, “something more distributed, more fluid and less rigid”.
Today, as a bioengineer at Stanford University in California, Boahen is among a small band of researchers trying to create this kind of computing by reverse-engineering the brain.
The brain is remarkably energy efficient and can carry out computations that challenge the world’s largest supercomputers, even though it relies on decidedly imperfect components: neurons that are a slow, variable, organic mess. Comprehending language, conducting abstract reasoning, controlling movement — the brain does all this and more in a package that is smaller than a shoebox, consumes less power than a household light bulb, and contains nothing remotely like a central processor.
To achieve similar feats in silicon, researchers are building systems of non-digital chips that function as much as possible like networks of real neurons. Just a few years ago, Boahen completed a device called Neurogrid that emulates a million neurons — about as many as there are in a honeybee’s brain. And now, after a quarter-century of development, applications for ‘neuromorphic technology’ are finally in sight. The technique holds promise for anything that needs to be small and run on low power, from smartphones and robots to artificial eyes and ears. That prospect has attracted many investigators to the field during the past five years, along with hundreds of millions of dollars in research funding from agencies in both the United States and Europe.
Neuromorphic devices are also providing neuroscientists with a powerful research tool, says Giacomo Indiveri at the Institute of Neuroinformatics (INI) in Zurich, Switzerland. By seeing which models of neural function do or do not work as expected in real physical systems, he says, “you get insight into why the brain is built the way it is”.
And, says Boahen, the neuromorphic approach should help to circumvent a looming limitation to Moore’s law — the longstanding trend of computer-chip manufacturers managing to double the number of transistors they can fit into a given space every two years or so. This relentless shrinkage will soon lead to the creation of silicon circuits so small and tightly packed that they no longer generate clean signals: electrons will leak through the components, making them as messy as neurons. Some researchers are aiming to solve this problem with software fixes, for example by using statistical error-correction techniques similar to those that help the Internet to run smoothly. But ultimately, argues Boahen, the most effective solution is the same one the brain arrived at millions of years ago.
“My goal is a new computing paradigm,” Boahen says, “something that will compute even when the components are too small to be reliable.”
The neuromorphic idea goes back to the 1980s and Carver Mead: a world-renowned pioneer in microchip design at the California Institute of Technology in Pasadena. He coined the term and was one of the first to emphasize the brain’s huge energy-efficiency advantage. “That’s been the fascination for me,” he says, “how in the heck can the brain do what it does?”
Mead’s strategy for answering that question was to mimic the brain’s low-power processing with ‘sub-threshold’ silicon: circuitry that operates at voltages too small to flip a standard computer bit from a 0 to a 1. At those voltages, there is still a tiny, irregular trickle of electrons running through the transistors — a spontaneous ebb and flow of current that is remarkably similar in size and variability to that carried by ions flowing through a channel in a neuron. With the addition of microscopic capacitors, resistors and other components to control these currents, Mead reasoned, it should be possible to make tiny circuits that exhibit the same electrical behavior as real neurons. They could be linked up in decentralized networks that function much like real neural circuits in the brain, with communication lines running between components rather than through a central processor.
By the 1990s, Mead and his colleagues had shown it was possible to build a realistic silicon neuron (see ‘Biological inspiration’). That device could accept outside electrical input through junctions that performed the role of synapses, the tiny structures through which nerve impulses jump from one neuron to the next. It allowed the incoming signals to build up voltage in the circuit’s interior, much as they do in real neurons. And if the accumulating voltage passed a certain threshold, the silicon neuron ‘fired’, producing a series of voltage spikes that traveled along a wire playing the part of an axon, the neuron’s communication cable. Although the spikes were ‘digital’ in the sense that they were either on or off, the body of the silicon neuron operated — like real neurons — in a non-digital way, meaning that the voltages and currents weren’t restricted to a few discrete values as they are in conventional chips.
That behavior mimics one key to the brain’s low-power usage: just like their biological counterparts, the silicon neurons simply integrated inputs, using very little energy, until they fired. By contrast, a conventional computer needs a constant flow of energy to run an internal clock, whether or not the chips are computing anything.
Mead’s group also demonstrated decentralized neural circuits — most notably in a silicon version of the eye’s retina. That device captured light using a 50-by-50 grid of detectors. When their activity was displayed on a computer screen, these silicon cells showed much the same response as their real counterparts to light, shadow and motion. Like the brain, this device saves energy by sending only the data that matters: most of the cells in the retina don’t fire until the light level changes. This has the effect of highlighting the edges of moving objects, while minimizing the amount of data that has to be transmitted and processed.
In those early days, researchers had their hands full mastering single-chip devices such as the silicon retina, says Boahen, who joined Mead’s lab in 1990. But by the end of the 1990s, he says, “we wanted to build a brain, and for that we needed large-scale communication”. That was a huge challenge: the standard coding algorithms for chip-to-chip communication had been devised for precisely coordinated digital signals, and wouldn’t work for the more-random spikes created by neuromorphic systems. Only in the 2000s did Boahen and others devise circuitry and algorithms that would work in this messier system, opening the way for a flurry of development in large-scale neuromorphic systems.
Among the first applications were large-scale emulators to give neuroscientists an easy way to test models of brain function. In September 2006, for example, Boahen launched the Neurogrid project: an effort to emulate a million neurons. That is only a tiny chunk of the 86 billion neurons in the human brain, but enough to model several of the densely interconnected columns of neurons thought to form the computational units of the human cortex. Neuroscientists can program Neurogrid to emulate almost any model of the cortex, says Boahen. They can then watch their model run at the same speed as the brain — hundreds to thousands of times faster than a conventional digital simulation. Graduate students and researchers have used it to test theoretical models of neural function for processes such as working memory, decision-making and visual attention.
“In terms of real efficiency, in terms of fidelity to the brain’s neuronal networks, Kwabena’s Neurogrid is well in advance of other large-scale neuromorphic systems,” says Rodney Douglas, co-founder of the INI and co-developer of the silicon neuron.
Read more . . .
The Latest on: Neuroelectronics
What makes memories stronger? - EurekAlert
April 29, 2019 - What makes memories stronger? EurekAlertA team of scientists at NeuroElectronics Research Flanders (NERF- empowered by imec, KU Leuven and VIB) found that highly demanding and rewarding ...
Important Memories Stored And Strengthened - - Today News Air
April 30, 2019 - Important Memories Stored And Strengthened - Today News AirImportant memories stored and strengthened. A group of scientists at NeuroElectronics Research Flanders discovered that extremely challenging.
From spinal cord injury to recovery: An essential role for neuronal feedback from below the site of injury - Science Daily
April 8, 2019 - From spinal cord injury to recovery: An essential role for neuronal feedback from below the site of injury Science DailySpinal cord injury disconnects communication between the brain and the spinal cord, disrupting control over part of the body. Studying the mechanisms of ...
Tiny implantable wireless devices could help people repair nerves and lose weight - Science Magazine
December 12, 2018 - Tiny implantable wireless devices could help people repair nerves and lose weight Science MagazineSome of the devices even dissolve in the body once their work is done.
Imaging whole brains of mice using functional ultrasound - DTR - Drug Target Review
December 5, 2018 - Imaging whole brains of mice using functional ultrasound - DTR Drug Target ReviewFunctional ultrasound imaging has been used by researchers to obtain high-quality, high-resolution images of whole brains of mice.
Printing microelectrode array sensors on gummi candy - Phys.Org
June 22, 2018 - Printing microelectrode array sensors on gummi candy Phys.OrgMicroelectrodes can be used for direct measurement of electrical signals in the brain or heart. These applications require soft materials, however. With existing ...
Sensors Printed Onto Gummy Candies Could Monitor Heart or Brain Functions - Interesting Engineering
June 23, 2018 - Sensors Printed Onto Gummy Candies Could Monitor Heart or Brain Functions Interesting EngineeringWhat if patients could swallow their own sensors? That's the question a team of European researchers answered by developing a new way to monitor patients.
Artificial neurons compute faster than the human brain - Nature.com
January 26, 2018 - Artificial neurons compute faster than the human brain Nature.comA computing system that mimics neural processing could make artificial intelligence more efficient — and more human.
Researchers Just Printed Sensors Onto Gummi Candies - Forbes
June 21, 2018 - Researchers Just Printed Sensors Onto Gummi Candies ForbesBut why? Read on. Gummi candies are known worldwide for their sweet taste, but researchers led by Neuroelectronics Professor Bernhard Wolfum at the ...
Fast, flexible ionic transistors for bioelectronic devices - Tech Xplore
February 27, 2019 - Fast, flexible ionic transistors for bioelectronic devices Tech XploreIGT-based NAND and NOR gates conform to the surface of orchid petals (left). Scale bar, 1cm. Optical micrographs of NOR (upper right) and NAND (lower right) ...
How your moving brain sees the world - Science Daily
November 20, 2018 - How your moving brain sees the world Science DailyWhat we see is not only determined by what is really there, but also depends on whether we are paying attention, whether we are moving, excited or interested.
Secreted amyloid-β precursor protein functions as a GABABR1a ligand to modulate synaptic transmission - Science Magazine
April 9, 2019 - Secreted amyloid-β precursor protein functions as a GABABR1a ligand to modulate synaptic transmission Science MagazineAlthough the pathological role of the amyloid-β precursor protein (APP) in Alzheimer's disease is well studied, the physiological role of this protein has remained ...
Brain-Machine Interfaces Make for Tricky Ethics - Wired News
February 25, 2009 - Brain-Machine Interfaces Make for Tricky Ethics Wired NewsDirect connections from brains to computers may someday help free paralyzed people from the constraints of their bodies. They're already used to reverse ...
Microelectrode arrays printed on gelatin and other soft materials - eeNews Europe
June 21, 2018 - Microelectrode arrays printed on gelatin and other soft materials eeNews EuropeA joint research team of the Technical University of Munich (TUM) and the Research Centre Jülich has printed patterns onto gummy bears. A joke of ...
A New Supercomputer Is the World's Fastest Brain-Mimicking Machine - Scientific American
November 5, 2018 - A New Supercomputer Is the World's Fastest Brain-Mimicking Machine Scientific AmericanScientific American is the essential guide to the most awe-inspiring advances in science and technology, explaining how they change our understanding of the ...
New brain region identified in spatial memory, navigation - UPI.com
August 16, 2017 - New brain region identified in spatial memory, navigation UPI.comResearchers at the Flanders Institute for Biotechnology have identified a new region of the brain responsible for navigation and spatial memory.
Neurotechnology provides real-time readouts of where rats think they are - Science Daily
December 4, 2018 - Neurotechnology provides real-time readouts of where rats think they are Science DailyScientists have demonstrated a new neurotechnology for reading out neural signals of position in real-time as rats run a maze, or replay it during sleep, with a ...
A Smarter Way for Learning Pros to Embrace Emerging Tech - ATD
March 27, 2019 - A Smarter Way for Learning Pros to Embrace Emerging Tech ATDThe new BUILDS framework can help L&D pros anticipate and evaluate emerging learning technologies.
Real-time readouts of thinking in rats - MIT News
December 19, 2018 - Real-time readouts of thinking in rats MIT NewsAn open-source system provides fast, accurate neural decoding and real-time readouts of where rats think they are. The system's development was led by ...
Research on largest network of cortical neurons to date: Robust network of connections between neurons performing similar tasks shows fundamentals of how brain circuits are wired - Science Daily
March 28, 2016 - Research on largest network of cortical neurons to date: Robust network of connections between neurons performing similar tasks shows fundamentals of how brain circuits are wired Science DailyThe largest network to date of connections between neurons in the cortex, where high-level processing occurs, has been published. It has revealed several ...
Graphene applications for bioelectronics and neuroprosthetics - Nanowerk
September 3, 2018 - Graphene applications for bioelectronics and neuroprosthetics NanowerkGraphene bioelectronics has become a ground-breaking field that offers exciting opportunities for developing new kinds of sensors capable of establishing ...
Imec Adds 4000-m2 Cleanroom for Integrated Circuit Production - Photonics.com
March 16, 2016 - Imec Adds 4000-m2 Cleanroom for Integrated Circuit Production Photonics.comLEUVEN, Belgium, March 16, 2016 — Nanoelectronics research center imec has announced the opening of a 4000-m2 cleanroom within its research facility, ...
12 technologies that can extend Moore's Law - IDG Connect
February 13, 2018 - 12 technologies that can extend Moore's Law IDG ConnectToday's technology industry was built upon the idea of continually improving hardware. After forty years, Moore's Law – the premise that the size of transistors ...
Peering into fish brains to see how they work - Science Daily
September 9, 2015 - Peering into fish brains to see how they work Science DailyOne of the fundamental challenges facing neuroscientists who want to understand how the brain works is actually figuring out how the brain is wired together ...
Imec cleanroom pushes towards 'the ultimate IC technology' - Electronics Weekly
March 14, 2016 - Imec cleanroom pushes towards 'the ultimate IC technology' Electronics WeeklyImec has opened a 4000 square-metre 300mm cleanroom which brings its total clean room capacity to 12000 square metres.
How to Hack the Hackers: The Human Side of Cyber Crime - Scientific American
May 11, 2016 - How to Hack the Hackers: The Human Side of Cyber Crime Scientific AmericanSay what you will about cybercriminals, says Angela Sasse, “their victims rave about the customer *service*”. Sasse is talking about ransomware: an extortion ...
Imitating synapses of the human brain could lead to smarter electronics - Phys.Org
November 11, 2015 - Imitating synapses of the human brain could lead to smarter electronics Phys.OrgMaking a computer that learns and remembers like a human brain is a daunting challenge. The complex organ has 86 billion neurons and trillions of ...
Imec officially opens new cleanroom - evertiq.com
March 11, 2016 - Imec officially opens new cleanroom evertiq.comImec has officially opened its new 300 mm cleanroom. With this 4'000 square-meter new facility, imec's semiconductor research cleanrooms now totals 12'000 ...
Alternative Fusion Technologies Heat Up - Scientific American
July 23, 2014 - Alternative Fusion Technologies Heat Up Scientific AmericanScientific American is the essential guide to the most awe-inspiring advances in science and technology, explaining how they change our understanding of the ...
Understanding Brain Functions Using Upconversion Nanoparticles - Photonics Online
February 14, 2018 - Understanding Brain Functions Using Upconversion Nanoparticles Photonics OnlineResearchers can now send light deep into the brain to study neural activities. Scientists from the National University of Singapore (NUS) have worked with an ...
Massive Open Online Courses, aka MOOCs, Transform Higher Education and Science - Scientific American
March 13, 2013 - Massive Open Online Courses, aka MOOCs, Transform Higher Education and Science Scientific AmericanScientific American is the essential guide to the most awe-inspiring advances in science and technology, explaining how they change our understanding of the ...
Science 2.0: Great New Tool, or Great Risk? - Scientific American
January 9, 2008 - Science 2.0: Great New Tool, or Great Risk? Scientific AmericanScientific American is the essential guide to the most awe-inspiring advances in science and technology, explaining how they change our understanding of the ...
Making a Career in Medical Engineering - Greater Kashmir
April 25, 2016 - Making a Career in Medical Engineering Greater KashmirProminent biomedical engineering applications include the development of biocompatible prostheses, various diagnostic and therapeutic medical devices ...