Neural networks could be implemented more quickly using new photonic technology
“Deep learning” computer systems, based on artificial neural networks that mimic the way the brain learns from an accumulation of examples, have become a hot topic in computer science. In addition to enabling technologies such as face- and voice-recognition software, these systems could scour vast amounts of medical data to find patterns that could be useful diagnostically, or scan chemical formulas for possible new pharmaceuticals.
But the computations these systems must carry out are highly complex and demanding, even for the most powerful computers.
Now, a team of researchers at MIT and elsewhere has developed a new approach to such computations, using light instead of electricity, which they say could vastly improve the speed and efficiency of certain deep learning computations. Their results appear today in the journal Nature Photonics in a paper by MIT postdoc Yichen Shen, graduate student Nicholas Harris, professors Marin Solja?i? and Dirk Englund, and eight others.
Solja?i? says that many researchers over the years have made claims about optics-based computers, but that “people dramatically over-promised, and it backfired.” While many proposed uses of such photonic computers turned out not to be practical, a light-based neural-network system developed by this team “may be applicable for deep-learning for some applications,” he says.
Traditional computer architectures are not very efficient when it comes to the kinds of calculations needed for certain important neural-network tasks. Such tasks typically involve repeated multiplications of matrices, which can be very computationally intensive in conventional CPU or GPU chips.
After years of research, the MIT team has come up with a way of performing these operations optically instead. “This chip, once you tune it, can carry out matrix multiplication with, in principle, zero energy, almost instantly,” Solja?i? says. “We’ve demonstrated the crucial building blocks but not yet the full system.”
By way of analogy, Solja?i? points out that even an ordinary eyeglass lens carries out a complex calculation (the so-called Fourier transform) on the light waves that pass through it. The way light beams carry out computations in the new photonic chips is far more general but has a similar underlying principle. The new approach uses multiple light beams directed in such a way that their waves interact with each other, producing interference patterns that convey the result of the intended operation. The resulting device is something the researchers call a programmable nanophotonic processor.
The result, Shen says, is that the optical chips using this architecture could, in principle, carry out calculations performed in typical artificial intelligence algorithms much faster and using less than one-thousandth as much energy per operation as conventional electronic chips. “The natural advantage of using light to do matrix multiplication plays a big part in the speed up and power savings, because dense matrix multiplications are the most power hungry and time consuming part in AI algorithms” he says.
The new programmable nanophotonic processor, which was developed in the Englund lab by Harris and collaborators, uses an array of waveguides that are interconnected in a way that can be modified as needed, programming that set of beams for a specific computation. “You can program in any matrix operation,” Harris says. The processor guides light through a series of coupled photonic waveguides. The team’s full proposal calls for interleaved layers of devices that apply an operation called a nonlinear activation function, in analogy with the operation of neurons in the brain.
To demonstrate the concept, the team set the programmable nanophotonic processor to implement a neural network that recognizes four basic vowel sounds. Even with this rudimentary system, they were able to achieve a 77 percent accuracy level, compared to about 90 percent for conventional systems. There are “no substantial obstacles” to scaling up the system for greater accuracy, Solja?i? says.
Englund adds that the programmable nanophotonic processor could have other applications as well, including signal processing for data transmission. “High-speed analog signal processing is something this could manage” faster than other approaches that first convert the signal to digital form, since light is an inherently analog medium. “This approach could do processing directly in the analog domain,” he says.
The team says it will still take a lot more effort and time to make this system useful; however, once the system is scaled up and fully functioning, it can find many user cases, such as data centers or security systems. The system could also be a boon for self-driving cars or drones, says Harris, or “whenever you need to do a lot of computation but you don’t have a lot of power or time.”
Learn more: New system allows optical “deep learning”
The Latest on: Neural networks
- Bringing neural networks to cellphones on July 22, 2017 at 12:05 am
MIT researchers have developed a tool for modeling neural networks, which leads to methods for reducing their energy consumption by 73 percent — making it feasible for them to run on cellphones rather than in the cloud. […]
- Intel Movidius: What Is a Neural Compute Stick and Why Would You Want One? on July 22, 2017 at 12:03 am
The Movidius Neural Compute Stick uses deep neural network processing and machine vision technology to "reduce barriers to developing, tuning and deploying AI [artificial intelligence] applications," Intel said in a statement yesterday. It called the ... […]
- Neural nets model audience reactions to movies on July 21, 2017 at 11:08 pm
"Our research shows that deep learning techniques, which use neural networks and have revolutionized the field of artificial intelligence, are effective at reducing data while capturing its hidden patterns." The research team applied FVAEs to 150 showings ... […]
- Intel forays into deep-learning arena; launches Movidius Neural Compute Stick on July 21, 2017 at 8:20 pm
accelerator that delivers dedicated deep neural network processing capabilities to a wide range of host devices at the edge. Designed for product developers and researchers, the Movidius Neural Compute Stick aims to reduce barriers to developing ... […]
- Neural Network Cross Entropy Error using Python on July 21, 2017 at 2:11 pm
I wrote an article in the July 2017 issue of Visual Studio Magazine titled “Neural Network Cross Entropy Error using Python”. See https://visualstudiomagazine.com/articles/2017/07/01/cross-entropy.aspx. For beginners to neural networks, cross entropy e ... […]
- Intel Launches Movidius Neural Compute Stick: Deep Learning and AI on a $79 USB Stick on July 20, 2017 at 9:00 am
As Intel states, the Movidius NCS is “the world’s first self-contained AI accelerator in a USB format,” and is designed to allow host devices to process deep neural networks natively – or in other words, at the edge. In turn, this provides ... […]
- Researchers Develop Neural Network Chip for Smartphones on July 20, 2017 at 8:56 am
MIT associate professor of electrical engineering and computer science Vivienne Sze and colleagues have introduced a computer chip for the software systems called neural networks. The networks are designed to power artificial intelligence (AI) systems that ... […]
- This MIT neural network translates pictures of food into recipes on July 20, 2017 at 1:59 am
Researchers from the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) have developed a neural network that can, in theory, can “look” at an image and find the recipe for the food that’s depicted. It’s the latest addition to an ... […]
- MIT Research Could Make Neural Networks Portable on July 19, 2017 at 4:30 am
Computers aren’t just getting faster these days–they’re getting smarter. That’s thanks largely to artificial neural networks, which consist of multiple nodes of processing designed to interpret and understand data more like a biological brain. […]
- Method for modeling neural networks' power consumption could help make the systems portable on July 18, 2017 at 11:11 pm
MIT researchers have designed new methods for paring down neural networks so that they’ll run more efficiently on handheld devices. Credit: Jose-Luis Olivares/MIT In recent years, the best-performing artificial-intelligence systems—in areas such as ... […]
via Google News and Bing News