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
Movies, Neural Networks Boost AI Language Skills
on July 18, 2018 at 7:52 pm
When we discuss about artificial intelligence (AI), how are machines learning? What kinds of projects feed into greater understanding? For our friends over at IBM, one surprising answer is movies. To ... […]
Google AI launches Quick Draw-like WeChat game to showcase neural networks in China
on July 18, 2018 at 4:32 pm
Over the past year, Google has been slowly returning to China via partnerships, investments, and apps. The latest is not a standalone application, but rather a drawing mini game for WeChat where machi... […]
Always Learning, Always Growing: How Neural Networks Do The Hard Work
on July 17, 2018 at 3:08 pm
he year 1958: Dwight D. Eisenhower was in the White House, Buddy Holly’s “Peggy Sue” ruled the airwaves, and a psychologist named Frank Rosenblatt was putting the finishing touches on a machine called ... […]
Neural Networks or Deep Learning in Natural Language Processing
on July 17, 2018 at 10:51 am
In last article i talked about challenges in natural language processing. Deep Learning is a branch of Machine Learning. It is just a fancy name for Neural Networks. Neural Networks is a machine learn... […]
Team develops new generation of artificial neural networks able to predict properties of organic compounds
on July 17, 2018 at 3:08 am
Scientists from Russia, Estonia and the United Kingdom have created a new method for predicting the bioconcentration factor (BCF) of organic molecules. Leveraging the classical models of physicochemic... […]
Artificial neural networks now able to help reveal a brain's structure
on July 17, 2018 at 2:54 am
Flood-filling network (FFN) neuron reconstructions in a songbird basal ganglia SBEM data set. Credit: MPI of Neurobiology/ Kornfeld The function of the brain is based on the connections between nerve ... […]
Neural Network From Scratch — Tumour Diagnosis
on July 14, 2018 at 7:34 pm
This article is written to illustraste how to build a simple 2 layer neural network from scratch! I have wanted to do this for a long time now as I noticed that when I was learning about neural networ... […]
Neural Networks And EPS Prediction
on July 13, 2018 at 7:27 am
Yes, it sounds a bit like the phrase "jumbo shrimp," in terms of sense, but a "long short-term memory neural network" is an important recent advance in artificial intelligence research. The term refer... […]
Artificial neural network made out of DNA
on July 7, 2018 at 12:46 pm
Researchers have constructed an artificial neural network from DNA. The network can be used to recognize highly complex and distorted molecular data. In what is a biological breakthrough, researchers ... […]
DNA-Based AI Neural Network Can Read Human Handwritten Numbers
on July 6, 2018 at 7:08 am
Researchers from Caltech have taught an artificial neural network made out of DNA to identify handwritten numbers. This brings artificial intelligence (AI) one step closer to integrating AI into synth... […]
via Google News and Bing News