Tag Archives: top
Move over, deep learning. Neuromorphic computing—the next big thing in artificial intelligence—is on fire.
Just last week, two studies individually unveiled computer chips modeled after information processing in the human brain.
The first, published in Nature Materials, found a perfect solution to deal with unpredictability at synapses—the gap between two neurons that transmit and store information. The second, published in Science Advances, further amped up the system’s computational power, filling synapses with nanoclusters of supermagnetic material to bolster information encoding.
The result? Brain-like hardware systems that compute faster—and more efficiently—than the human brain.
“Ultimately we want a chip as big as a fingernail to replace one big supercomputer,” said Dr. Jeehwan Kim, who led the first study at MIT in Cambridge, Massachusetts.
Experts are hopeful.
“The field’s full of hype, and it’s nice to see quality work presented in an objective way,” said Dr. Carver Mead, an engineer at the California Institute of Technology in Pasadena not involved in the work.
Software to Hardware
The human brain is the ultimate computational wizard. With roughly 100 billion neurons densely packed into the size of a small football, the brain can deftly handle complex computation at lightning speed using very little energy.
AI experts have taken note. The past few years saw brain-inspired algorithms that can identify faces, falsify voices, and play a variety of games at—and often above—human capability.
But software is only part of the equation. Our current computers, with their transistors and binary digital systems, aren’t equipped to run these powerful algorithms.
That’s where neuromorphic computing comes in. The idea is simple: fabricate a computer chip that mimics the brain at the hardware level. Here, data is both processed and stored within the chip in an analog manner. Each artificial synapse can accumulate and integrate small bits of information from multiple sources and fire only when it reaches a threshold—much like its biological counterpart.
Experts believe the speed and efficiency gains will be enormous.
For one, the chips will no longer have to transfer data between the central processing unit (CPU) and storage blocks, which wastes both time and energy. For another, like biological neural networks, neuromorphic devices can support neurons that run millions of streams of parallel computation.
Optimism aside, reproducing the biological synapse in hardware form hasn’t been as easy as anticipated.
Neuromorphic chips exist in many forms, but often look like a nanoscale metal sandwich. The “bread” pieces are generally made of conductive plates surrounding a switching medium—a conductive material of sorts that acts like the gap in a biological synapse.
When a voltage is applied, as in the case of data input, ions move within the switching medium, which then creates conductive streams to stimulate the downstream plate. This change in conductivity mimics the way biological neurons change their “weight,” or the strength of connectivity between two adjacent neurons.
But so far, neuromorphic synapses have been rather unpredictable. According to Kim, that’s because the switching medium is often comprised of material that can’t channel ions to exact locations on the downstream plate.
“Once you apply some voltage to represent some data with your artificial neuron, you have to erase and be able to write it again in the exact same way,” explains Kim. “But in an amorphous solid, when you write again, the ions go in different directions because there are lots of defects.”
In his new study, Kim and colleagues swapped the jelly-like switching medium for silicon, a material with only a single line of defects that acts like a channel to guide ions.
The chip starts with a thin wafer of silicon etched with a honeycomb-like pattern. On top is a layer of silicon germanium—something often present in transistors—in the same pattern. This creates a funnel-like dislocation, a kind of Grand Canal that perfectly shuttles ions across the artificial synapse.
The researchers then made a neuromorphic chip containing these synapses and shot an electrical zap through them. Incredibly, the synapses’ response varied by only four percent—much higher than any neuromorphic device made with an amorphous switching medium.
In a computer simulation, the team built a multi-layer artificial neural network using parameters measured from their device. After tens of thousands of training examples, their neural network correctly recognized samples 95 percent of the time, just 2 percent lower than state-of-the-art software algorithms.
The upside? The neuromorphic chip requires much less space than the hardware that runs deep learning algorithms. Forget supercomputers—these chips could one day run complex computations right on our handheld devices.
A Magnetic Boost
Meanwhile, in Boulder, Colorado, Dr. Michael Schneider at the National Institute of Standards and Technology also realized that the standard switching medium had to go.
“There must be a better way to do this, because nature has figured out a better way to do this,” he says.
His solution? Nanoclusters of magnetic manganese.
Schneider’s chip contained two slices of superconducting electrodes made out of niobium, which channel electricity with no resistance. When researchers applied different magnetic fields to the synapse, they could control the alignment of the manganese “filling.”
The switch gave the chip a double boost. For one, by aligning the switching medium, the team could predict the ion flow and boost uniformity. For another, the magnetic manganese itself adds computational power. The chip can now encode data in both the level of electrical input and the direction of the magnetisms without bulking up the synapse.
It seriously worked. At one billion times per second, the chips fired several orders of magnitude faster than human neurons. Plus, the chips required just one ten-thousandth of the energy used by their biological counterparts, all the while synthesizing input from nine different sources in an analog manner.
The Road Ahead
These studies show that we may be nearing a benchmark where artificial synapses match—or even outperform—their human inspiration.
But to Dr. Steven Furber, an expert in neuromorphic computing, we still have a ways before the chips go mainstream.
Many of the special materials used in these chips require specific temperatures, he says. Magnetic manganese chips, for example, require temperatures around absolute zero to operate, meaning they come with the need for giant cooling tanks filled with liquid helium—obviously not practical for everyday use.
Another is scalability. Millions of synapses are necessary before a neuromorphic device can be used to tackle everyday problems such as facial recognition. So far, no deal.
But these problems may in fact be a driving force for the entire field. Intense competition could push teams into exploring different ideas and solutions to similar problems, much like these two studies.
If so, future chips may come in diverse flavors. Similar to our vast array of deep learning algorithms and operating systems, the computer chips of the future may also vary depending on specific requirements and needs.
It is worth developing as many different technological approaches as possible, says Furber, especially as neuroscientists increasingly understand what makes our biological synapses—the ultimate inspiration—so amazingly efficient.
Image Credit: arakio / Shutterstock.com Continue reading
The first time Dr. Blake Richards heard about deep learning, he was convinced that he wasn’t just looking at a technique that would revolutionize artificial intelligence. He also knew he was looking at something fundamental about the human brain.
That was the early 2000s, and Richards was taking a course with Dr. Geoff Hinton at the University of Toronto. Hinton, a pioneer architect of the algorithm that would later take the world by storm, was offering an introductory course on his learning method inspired by the human brain.
The key words here are “inspired by.” Despite Richards’ conviction, the odds were stacked against him. The human brain, as it happens, seems to lack a critical function that’s programmed into deep learning algorithms. On the surface, the algorithms were violating basic biological facts already proven by neuroscientists.
But what if, superficial differences aside, deep learning and the brain are actually compatible?
Now, in a new study published in eLife, Richards, working with DeepMind, proposed a new algorithm based on the biological structure of neurons in the neocortex. Also known as the cortex, this outermost region of the brain is home to higher cognitive functions such as reasoning, prediction, and flexible thought.
The team networked their artificial neurons together into a multi-layered network and challenged it with a classic computer vision task—identifying hand-written numbers.
The new algorithm performed well. But the kicker is that it analyzed the learning examples in a way that’s characteristic of deep learning algorithms, even though it was completely based on the brain’s fundamental biology.
“Deep learning is possible in a biological framework,” concludes the team.
Because the model is only a computer simulation at this point, Richards hopes to pass the baton to experimental neuroscientists, who could actively test whether the algorithm operates in an actual brain.
If so, the data could then be passed back to computer scientists to work out the next generation of massively parallel and low-energy algorithms to power our machines.
It’s a first step towards merging the two fields back into a “virtuous circle” of discovery and innovation.
The blame game
While you’ve probably heard of deep learning’s recent wins against humans in the game of Go, you might not know the nitty-gritty behind the algorithm’s operations.
In a nutshell, deep learning relies on an artificial neural network with virtual “neurons.” Like a towering skyscraper, the network is structured into hierarchies: lower-level neurons process aspects of an input—for example, a horizontal or vertical stroke that eventually forms the number four—whereas higher-level neurons extract more abstract aspects of the number four.
To teach the network, you give it examples of what you’re looking for. The signal propagates forward in the network (like climbing up a building), where each neuron works to fish out something fundamental about the number four.
Like children trying to learn a skill the first time, initially the network doesn’t do so well. It spits out what it thinks a universal number four should look like—think a Picasso-esque rendition.
But here’s where the learning occurs: the algorithm compares the output with the ideal output, and computes the difference between the two (dubbed “error”). This error is then “backpropagated” throughout the entire network, telling each neuron: hey, this is how far off you were, so try adjusting your computation closer to the ideal.
Millions of examples and tweakings later, the network inches closer to the desired output and becomes highly proficient at the trained task.
This error signal is crucial for learning. Without efficient “backprop,” the network doesn’t know which of its neurons are off kilter. By assigning blame, the AI can better itself.
The brain does this too. How? We have no clue.
What’s clear, though, is that the deep learning solution doesn’t work.
Backprop is a pretty needy function. It requires a very specific infrastructure for it to work as expected.
For one, each neuron in the network has to receive the error feedback. But in the brain, neurons are only connected to a few downstream partners (if that). For backprop to work in the brain, early-level neurons need to be able to receive information from billions of connections in their downstream circuits—a biological impossibility.
And while certain deep learning algorithms adapt a more local form of backprop— essentially between neurons—it requires their connection forwards and backwards to be symmetric. This hardly ever occurs in the brain’s synapses.
More recent algorithms adapt a slightly different strategy, in that they implement a separate feedback pathway that helps the neurons to figure out errors locally. While it’s more biologically plausible, the brain doesn’t have a separate computational network dedicated to the blame game.
What it does have are neurons with intricate structures, unlike the uniform “balls” that are currently applied in deep learning.
The team took inspiration from pyramidal cells that populate the human cortex.
“Most of these neurons are shaped like trees, with ‘roots’ deep in the brain and ‘branches’ close to the surface,” says Richards. “What’s interesting is that these roots receive a different set of inputs than the branches that are way up at the top of the tree.”
This is an illustration of a multi-compartment neural network model for deep learning. Left: Reconstruction of pyramidal neurons from mouse primary visual cortex. Right: Illustration of simplified pyramidal neuron models. Image Credit: CIFAR
Curiously, the structure of neurons often turn out be “just right” for efficiently cracking a computational problem. Take the processing of sensations: the bottoms of pyramidal neurons are right smack where they need to be to receive sensory input, whereas the tops are conveniently placed to transmit feedback errors.
Could this intricate structure be evolution’s solution to channeling the error signal?
The team set up a multi-layered neural network based on previous algorithms. But rather than having uniform neurons, they gave those in middle layers—sandwiched between the input and output—compartments, just like real neurons.
When trained with hand-written digits, the algorithm performed much better than a single-layered network, despite lacking a way to perform classical backprop. The cell-like structure itself was sufficient to assign error: the error signals at one end of the neuron are naturally kept separate from input at the other end.
Then, at the right moment, the neuron brings both sources of information together to find the best solution.
There’s some biological evidence for this: neuroscientists have long known that the neuron’s input branches perform local computations, which can be integrated with signals that propagate backwards from the so-called output branch.
However, we don’t yet know if this is the brain’s way of dealing blame—a question that Richards urges neuroscientists to test out.
What’s more, the network parsed the problem in a way eerily similar to traditional deep learning algorithms: it took advantage of its multi-layered structure to extract progressively more abstract “ideas” about each number.
“[This is] the hallmark of deep learning,” the authors explain.
The Deep Learning Brain
Without doubt, there will be more twists and turns to the story as computer scientists incorporate more biological details into AI algorithms.
One aspect that Richards and team are already eyeing is a top-down predictive function, in which signals from higher levels directly influence how lower levels respond to input.
Feedback from upper levels doesn’t just provide error signals; it could also be nudging lower processing neurons towards a “better” activity pattern in real-time, says Richards.
The network doesn’t yet outperform other non-biologically derived (but “brain-inspired”) deep networks. But that’s not the point.
“Deep learning has had a huge impact on AI, but, to date, its impact on neuroscience has been limited,” the authors say.
Now neuroscientists have a lead they could experimentally test: that the structure of neurons underlie nature’s own deep learning algorithm.
“What we might see in the next decade or so is a real virtuous cycle of research between neuroscience and AI, where neuroscience discoveries help us to develop new AI and AI can help us interpret and understand our experimental data in neuroscience,” says Richards.
Image Credit: christitzeimaging.com / Shutterstock.com Continue reading
I don’t have to open the doors of AImotive’s white 2015 Prius to see that it’s not your average car. This particular Prius has been christened El Capitan, the name written below the rear doors, and two small cameras are mounted on top of the car. Bundles of wire snake out from them, as well as from the two additional cameras on the car’s hood and trunk.
Inside is where things really get interesting, though. The trunk holds a computer the size of a microwave, and a large monitor covers the passenger glove compartment and dashboard. The center console has three switches labeled “Allowed,” “Error,” and “Active.”
Budapest-based AImotive is working to provide scalable self-driving technology alongside big players like Waymo and Uber in the autonomous vehicle world. On a highway test ride with CEO Laszlo Kishonti near the company’s office in Mountain View, California, I got a glimpse of just how complex that world is.
Camera-Based Feedback System
AImotive’s approach to autonomous driving is a little different from that of some of the best-known systems. For starters, they’re using cameras, not lidar, as primary sensors. “The traffic system is visual and the cost of cameras is low,” Kishonti said. “A lidar can recognize when there are people near the car, but a camera can differentiate between, say, an elderly person and a child. Lidar’s resolution isn’t high enough to recognize the subtle differences of urban driving.”
Image Credit: AImotive
The company’s aiDrive software uses data from the camera sensors to feed information to its algorithms for hierarchical decision-making, grouped under four concurrent activities: recognition, location, motion, and control.
Kishonti pointed out that lidar has already gotten more cost-efficient, and will only continue to do so.
“Ten years ago, lidar was best because there wasn’t enough processing power to do all the calculations by AI. But the cost of running AI is decreasing,” he said. “In our approach, computer vision and AI processing are key, and for safety, we’ll have fallback sensors like radar or lidar.”
aiDrive currently runs on Nvidia chips, which Kishonti noted were originally designed for graphics, and are not terribly efficient given how power-hungry they are. “We’re planning to substitute lower-cost, lower-energy chips in the next six months,” he said.
Testing in Virtual Reality
Waymo recently announced its fleet has now driven four million miles autonomously. That’s a lot of miles, and hard to compete with. But AImotive isn’t trying to compete, at least not by logging more real-life test miles. Instead, the company is doing 90 percent of its testing in virtual reality. “This is what truly differentiates us from competitors,” Kishonti said.
He outlined the three main benefits of VR testing: it can simulate scenarios too dangerous for the real world (such as hitting something), too costly (not every company has Waymo’s funds to run hundreds of cars on real roads), or too time-consuming (like waiting for rain, snow, or other weather conditions to occur naturally and repeatedly).
“Real-world traffic testing is very skewed towards the boring miles,” he said. “What we want to do is test all the cases that are hard to solve.”
On a screen that looked not unlike multiple games of Mario Kart, he showed me the simulator. Cartoon cars cruised down winding streets, outfitted with all the real-world surroundings: people, trees, signs, other cars. As I watched, a furry kangaroo suddenly hopped across one screen. “Volvo had an issue in Australia,” Kishonti explained. “A kangaroo’s movement is different than other animals since it hops instead of running.” Talk about cases that are hard to solve.
AImotive is currently testing around 1,000 simulated scenarios every night, with a steadily-rising curve of successful tests. These scenarios are broken down into features, and the car’s behavior around those features fed into a neural network. As the algorithms learn more features, the level of complexity the vehicles can handle goes up.
On the Road
After Kishonti and his colleagues filled me in on the details of their product, it was time to test it out. A safety driver sat in the driver’s seat, a computer operator in the passenger seat, and Kishonti and I in back. The driver maintained full control of the car until we merged onto the highway. Then he flicked the “Allowed” switch, his copilot pressed the “Active” switch, and he took his hands off the wheel.
What happened next, you ask?
A few things. El Capitan was going exactly the speed limit—65 miles per hour—which meant all the other cars were passing us. When a car merged in front of us or cut us off, El Cap braked accordingly (if a little abruptly). The monitor displayed the feed from each of the car’s cameras, plus multiple data fields and a simulation where a blue line marked the center of the lane, measured by the cameras tracking the lane markings on either side.
I noticed El Cap wobbling out of our lane a bit, but it wasn’t until two things happened in a row that I felt a little nervous: first we went under a bridge, then a truck pulled up next to us, both bridge and truck casting a complete shadow over our car. At that point El Cap lost it, and we swerved haphazardly to the right, narrowly missing the truck’s rear wheels. The safety driver grabbed the steering wheel and took back control of the car.
What happened, Kishonti explained, was that the shadows made it hard for the car’s cameras to see the lane markings. This was a new scenario the algorithm hadn’t previously encountered. If we’d only gone under a bridge or only been next to the truck for a second, El Cap may not have had so much trouble, but the two events happening in a row really threw the car for a loop—almost literally.
“This is a new scenario we’ll add to our testing,” Kishonti said. He added that another way for the algorithm to handle this type of scenario, rather than basing its speed and positioning on the lane markings, is to mimic nearby cars. “The human eye would see that other cars are still moving at the same speed, even if it can’t see details of the road,” he said.
After another brief—and thankfully uneventful—hands-off cruise down the highway, the safety driver took over, exited the highway, and drove us back to the office.
Driving into the Future
I climbed out of the car feeling amazed not only that self-driving cars are possible, but that driving is possible at all. I squint when driving into a tunnel, swerve to avoid hitting a stray squirrel, and brake gradually at stop signs—all without consciously thinking to do so. On top of learning to steer, brake, and accelerate, self-driving software has to incorporate our brains’ and bodies’ unconscious (but crucial) reactions, like our pupils dilating to let in more light so we can see in a tunnel.
Despite all the progress of machine learning, artificial intelligence, and computing power, I have a wholly renewed appreciation for the thing that’s been in charge of driving up till now: the human brain.
Kishonti seemed to feel similarly. “I don’t think autonomous vehicles in the near future will be better than the best drivers,” he said. “But they’ll be better than the average driver. What we want to achieve is safe, good-quality driving for everyone, with scalability.”
AImotive is currently working with American tech firms and with car and truck manufacturers in Europe, China, and Japan.
Image Credit: Alex Oakenman / Shutterstock.com Continue reading
Bitcoin Is a Delusion That Could Conquer the WorldDerek Thompson | The Atlantic“What seems most certain is that the future of money will test our conventional definitions—of currencies, of bubbles, and of initial offerings. What’s happening this month with bitcoin feels like an unsustainable paroxysm. But it’s foolish to try to develop rational models for when such a market will correct itself. Prices, like currencies, are collective illusions.”
This Engineer Is Building a DIY Mars Habitat in His BackyardDaniel Oberhaus | Motherboard“For over a year, Raymond and his wife have been running a fully operational, self-sustaining ‘Mars habitat’ in their backyard. They’ve personally sunk around $200,000 into the project and anticipate spending several thousand more before they’re finished. The habitat is the subject of a popularYouTube channel maintained by Raymond, where he essentiallyLARPs the 2015 Matt Damon film The Martian for an audience of over 20,000 loyal followers.”
The FCC Just Voted to Repeal Its Net Neutrality Rules, in a Sweeping Act of DeregulationBrian Fung | The Washington Post“The 3-2 vote, which was along party lines, enabled the FCC’s Republican chairman, AjitPai, to follow through on his promise to repeal the government’s 2015 net neutrality rules, which required Internet providers to treat all websites, large and small, equally.”
Sexism’s National Reckoning and the Tech Women Who Blazed the TrailTekla S. Perry | IEEE Spectrum“Cassidy and other women in tech who spoke during the one-day event stressed that the watershed came not because women finally broke the silence about sexual harassment, whatever Time’s editors may believe. The change came because the women were finally listened to and the bad actors faced repercussions.”
These Technologies Will Shape the Future, According to One of Silicon Valley’s Top VC FirmsDaniel Terdiman | Fast Company“The question then, is what are the technologies that are going to drive the future. At Andreessen Horowitz, a picture of that future, at least the next 10 years or so, is coming into focus.During a recent firm summit, Evans laid out his vision for the most significant tech opportunities of the next decade.On the surface, the four areas he identifies–autonomy, mixed-reality, cryptocurrencies, and artificial intelligence–aren’t entirely surprises.”
Image Credit: Solfer / Shutterstock.com Continue reading