Featured Product
This Week in Quality Digest Live
Operations Features
Lisa Apolinski
Adding what customers want
Adam Zewe
Understanding how machine-learning models behave to apply them more broadly
Tony Schmitz
US manufacturing is high-tech and needs skilled workers
Sarah Burlingame
Coaching can keep management and employees on track
Del Williams
Options to address the risk of combustible dust explosions for NFPA 61 compliance

More Features

Operations News
Equator system aids manufacturers of precision firearm parts
Datanomix chosen for its No Operator Input approach to production monitoring and out-of-the-box data automation
Safe trading practices to secure supply chain activities
VSL hosts special edition of show at new center in Rotterdam
Latest line touts comprehensive coverage, ease of use
Same price, double the resolution and accuracy
Former service partner provides honing and deep-hole drilling solutions
Low-cost prevention of catastrophic failures

More News

Adam Zewe

Operations

Deep Learning With Light

Optics accelerate machine-learning computations on smart speakers and other low-power connected devices

Published: Wednesday, November 16, 2022 - 12:01

Ask a smart home device for the weather forecast, and it takes several seconds to respond. One reason for this latency is that connected devices don’t have enough memory or power to store and run the enormous machine-learning models needed for the device to understand what a user is asking of it. The model is stored in a data center that may be hundreds of miles away, where the answer is computed and sent to the device.

MIT researchers have created a new method for computing directly on these devices that drastically reduces this latency. Their technique shifts the memory-intensive steps of running a machine-learning model to a central server where components of the model are encoded onto light waves.

The waves are transmitted to a connected device using fiber optics, which enables tons of data to be sent lightning-fast through a network. The receiver then employs a simple optical device that rapidly performs computations using the parts of a model carried by those light waves.

This technique leads to more than a hundredfold improvement in energy efficiency compared to other methods. It could also improve security, since a user’s data don’t need to be transferred to a central location for computation.

This method could enable a self-driving car to make decisions in real time while using just a tiny percentage of the energy currently required by power-hungry computers. It could also allow a user to have a latency-free conversation with a smart home device, be used for live video processing over cellular networks, or even enable high-speed image classification on a spacecraft millions of miles from Earth.

“Every time you want to run a neural network, you have to run the program, and how fast you can run the program depends on how fast you can pipe the program in from memory,” says senior author Dirk Englund, an associate professor in the Department of Electrical Engineering and Computer Science (EECS) and member of the MIT Research Laboratory of Electronics. “Our pipe is massive—it corresponds to sending a full feature-length movie over the internet every millisecond or so. That is how fast data come into our system. And it can compute as fast as that.” 

Joining Englund on the paper are lead author and EECS grad student Alexander Sludds; EECS grad student Saumil Bandyopadhyay; research scientist Ryan Hamerly; and others from MIT, the MIT Lincoln Laboratory, and Nokia Corp. The research is published in the journal Science.

Lightening the load

Neural networks are machine-learning models that use layers of connected nodes, or neurons, to recognize patterns in datasets and perform tasks such as classifying images or recognizing speech. But these models can contain billions of weight parameters, which are numeric values that transform input data as they are processed. These weights must be stored in memory. At the same time, the data transformation process involves billions of algebraic computations that require a great deal of power to perform.

The process of fetching data (the weights of the neural network, in this case) from memory and moving them to the parts of a computer that do the actual computation is one of the biggest limiting factors to speed and energy efficiency, says Sludds.

“So our thought was, why don’t we take all that heavy lifting—the process of fetching billions of weights from memory—move it away from the edge device, and put it someplace where we have abundant access to power and memory, which gives us the ability to fetch those weights quickly,” he says.

The neural network architecture they developed, Netcast, involves storing weights in a central server that is connected to a novel piece of hardware called a smart transceiver. This smart transceiver, a thumb-sized chip that can receive and transmit data, uses technology known as silicon photonics to fetch trillions of weights from memory each second.

It receives weights as electrical signals and imprints them onto light waves. Since the weight data are encoded as bits (1s and 0s), the transceiver converts them by switching lasers; a laser is turned on for a 1 and off for a 0. It combines these light waves and then periodically transfers them through a fiberoptic network so a client device doesn’t need to query the server to receive them.

“Optics is great because there are many ways to carry data within optics,” explains Bandyopadhyay. “For instance, you can put data on different colors of light, and that enables a much higher data throughput and greater bandwidth than with electronics.” 

Trillions per second

Once the light waves arrive at the client device, a simple optical component known as a broadband “Mach-Zehnder” modulator uses them to perform super-fast analog computation. This involves encoding input data from the device, such as sensor information, onto the weights. Then it sends each individual wavelength to a receiver that detects the light and measures the result of the computation.

The researchers devised a way to use this modulator to do trillions of multiplications per second, which vastly increases the speed of computation on the device while using only a tiny amount of power.   

“In order to make something faster, you need to make it more energy efficient,” Sludds says. “But there is a trade-off. We’ve built a system that can operate with about a milliwatt of power but still do trillions of multiplications per second. In terms of both speed and energy efficiency, that is a gain of orders of magnitude.” 

They tested this architecture by sending weights over an 86-km fiber that connects their lab to MIT Lincoln Laboratory. Netcast enabled machine-learning with high accuracy—98.7 percent for image classification and 98.8 percent for digit recognition—at rapid speeds.

“We had to do some calibration, but I was surprised by how little work we had to do to achieve such high accuracy out of the box,” adds Hamerly. “We were able to get commercially relevant accuracy.”

Moving forward, the researchers want to iterate on the smart transceiver chip to achieve even better performance. They also want to miniaturize the receiver, which is currently the size of a shoebox, down to the size of a single chip so it could fit onto a smart device such as a cellphone.

“Using photonics and light as a platform for computing is a really exciting area of research with potentially huge implications on the speed and efficiency of our information technology landscape,” says Euan Allen, a Royal Academy of Engineering Research Fellow at the University of Bath, who was not involved with this work. “The work of Sludds et al is an exciting step toward seeing real-world implementations of such devices, introducing a new and practical edge-computing scheme whilst also exploring some of the fundamental limitations of computation at very low (single-photon) light levels.”

First published Oct. 20, 2022, on MIT News.

Discuss

About The Author

Adam Zewe’s picture

Adam Zewe

Adam Zewe is a writer for Massachusetts Institute of Technology, covering the electrical engineering and computer science beat in the MIT News Office.