Accelerating Optical Communications with AI

 Accelerating Optical Communications with AI     By Chris Edwards

Communications of the ACM, July 2023, Vol. 66 No. 7, Pages 13-15  10.1145/3595957

Photonic computing has seen its share of research breakthroughs and deep research winters, much like the history of artificial intelligence (AI). Now, with the resurgence of AI, the huge amounts of energy today's large neural-network models need when running on electronic computers is reawakening interest in uniting the two.

More than 30 years ago, during one of the booms in research into artificial neural networks, Demetri Psaltis and colleagues at the California Institute of Technology demonstrated how techniques from holography could perform rudimentary face recognition. The team members showed they could store as many as one billion weights for a two-layer neural network using the core elements from a liquid-crystal display. Similar spatial light modulators became the foundation of several attempts to commercialize optical computing technology, including those by U.K.-based startup Optalysys, which has focused in recent years on applying the technology to accelerating homomorphic encryption to support secure remote computing.

Though some groups are using spatial light modulators for AI, they represent just one category of optical computer suitable for the job. There are also decisions as to which form of neural networks best suits optical computing. Some techniques focus on the matrix-arithmetic operations of mainstream deep learning pipelines, while others focus squarely on emulating the spike trains of biological brains.

What all the proposed systems have in common is the possibility that, by using photons to communicate and calculate, they will deliver major advantages in density and speed over systems based purely on electrical signaling. A 2021 study of inferencing based on matrix arithmetic by Mitchell Nahmias, now CTO of startup Luminous Computing, and colleagues at Princeton University argued the theoretical efficiency of AI inferencing in the optical domain could far surpass that of conventional accelerators based on existing electronics-only architectures.

The key issue for machine learning is the amount of energy needed to move data around accelerators. Electronic accelerators often employ strategies to cache as much data as possible to reduce this overhead, with major trade-offs concerning whether temporary results or weights are held in the cache depending on the model's structure. However, the energy cost of delivering photons over distances larger than the span of a single chip is far lower than it is for electrical signaling.

A second potential advantage of photonic AI comes from the ease with which it can handle complex operations in the analog domain, though the power savings achievable here are less certain than for communications. Whereas matrix arithmetic relies on highly parallelized hardware circuits for performance in conventional systems, simply passing photons through an optical component such as a Mach-Zehnder interferometer (MZi) or micro-ring resonator will perform arithmetic requiring hundreds or even thousands of logic gates in a digital circuit.

In the MZi, coherent beams of light pass through a succession of couplers and phase shifters. At each coupling point, the interference between the beams results in phase shifts that can be interpreted as part of a matrix multiplication. A 4x4 matrix operation requires just four inputs that feed into six coupling elements, with four output ports providing the result. The speed of the operations is limited only by the rate at which coherent pulses can be passed through the array.

In analog architectures, noise presents a significant hurdle. Work by numerous groups on accelerating inference operations has shown that deep neural networks can work successfully at an effective resolution of 4 bits, but hardware overhead and energy rise quickly as resolution increases. These effects may limit the practical energy advantage of photonic designs.

Estimates by Alexander Tait, assistant professor at Queen's University, Ontario, found the potential power-savings easily eroded by the practical limitations of today's optical devices. Tait calculated just 500 micro-ring resonators acting as neurons in a fully connected layout could fit onto a single 1cm2 die using early 2020s technology. But operating at 10GHz, it would require a kilowatt of power. Tait stresses the example shows the impact of the current need for heaters to tune optical properties. Scaling and design changes could bring the energy down dramatically. "The heaters are certainly a solvable problem," he says.  .... ' 

Optical Algorithm Simplifies Analog AI Training

Thinking this, possibilities and  implications.  

Optical Algorithm Simplifies Analog AI Training When backpropagation loses its bite, this approach may substitute  ...    in IEEE Spectrum

Researchers have developed a range of analog and other unconventional machine learning systems in the expectation that they will prove vastly more energy efficient than today’s computers. But training these AIs to do their tasks has been a big stumbling block. Researchers at NTT Device Technology Labs and the University of Tokyo now say they’ve come up with a training algorithm (announced by NTT last month) that goes a long way toward letting these systems meet their promise.

Their results, established on an optical analog computer, represent progress towards obtaining the potential efficiency gains that researchers have long sought from “unconventional” computer architectures.

Modern AI programs use a biologically-inspired architecture called an artificial neural network to execute tasks like image recognition or text generation. The strength of connections between artificial neurons, which control the outputs of the computation, must be modified or trained using standard algorithms. The most prominent of these algorithms is called backpropagation, which updates the connection strengths to reduce the network’s errors, while it processes trial data. Because adjustments to some parameters depend on adjustments to others, there is a need for active information passing and routing by the computer.

As Spectrum has elsewhere explained, “Error backpropagation is like running inference in reverse, moving from the last layer of the network back to the first layer; weight update then combines information from the original forward inference run with these backpropagated errors to adjust the network weights in a way that makes the model more accurate.”

Alternative computing architectures, which trade complexity for efficiency, often cannot perform the information passing required by the algorithm. As a consequence, the trained parameters of the network must be obtained from an independent physics simulation of the entire hardware setup and its information processing. But creating simulations of sufficient quality can itself be challenging.  ... ' 

Breaking the Scaling Limits of Analog Optical Computing

 Reducing error in optical networks 

Breaking the Scaling Limits of Analog Computing   By MIT News, November 30, 2022

Using the new technique, the larger an optical neural network becomes, the lower the error in its computations. 

Massachusetts Institute of Technology researchers have developed a technique that greatly reduces the error in an optical neural network.

As machine-learning models become larger and more complex, they require faster and more energy-efficient hardware to perform computations. Conventional digital computers are struggling to keep up.

An analog optical neural network could perform the same tasks as a digital one, such as image classification or speech recognition, but because computations are performed using light instead of electrical signals, optical neural networks can run many times faster while consuming less energy.

However, these analog devices are prone to hardware errors that can make computations less precise. Microscopic imperfections in hardware components are one cause of these errors. In an optical neural network that has many connected components, errors can quickly accumulate.

Even with error-correction techniques, due to fundamental properties of the devices that make up an optical neural network, some amount of error is unavoidable. A network that is large enough to be implemented in the real world would be far too imprecise to be effective.

From MIT News   View Full Article     

Intel Announces Silicon Photonics

 Quite interesting from Intel

Intel announces silicon photonics advancement towards optical I/O

Arne Verheyde @witeken in Venturebeat

Intel has demonstrated an eight-wavelength laser array on a silicon wafer. The research paves the way for the next generation of integrated silicon photonics products in the data center, such as switches with co-packaged optics and chiplets for optical interconnects.

From discrete transceivers to integrated photonics  .... 

Intel has been selling 100G transceivers based on silicon photonics for about half a decade, and has also started to commercialize up to 400G transceivers, with 800G also on the roadmap. Gartner forecasts that by 2025 over 20% of all data center communication will be based on silicon photonics, up from 5% in 2020, representing a $2.6 billion market. 

However, to further reduce power, the next step for silicon photonics is to become more tightly integrated with other silicon. The first instantiation of this is co-packaged optics, expected in the next two years. The ultimate goal is fully integrated photonics, where the photonic integrated circuit is directly connected to the rest of the compute using advanced packaging. 

Integrated DWDM laser array

The latest research from Intel Labs delivers what it calls “industry-leading” advancements in multiwavelength integrated optics, Intel claims. It concerns the demonstration of an eight-wavelength distributed feedback (DFB) laser array that was fully integrated on a 300mm silicon wafer using Intel’s hybrid silicon photonics platform, paving the way for cost-effective high-volume manufacturing for broad deployment. The laser achieved power uniformity and wavelength spacing uniformity that exceed industry specifications.

One of the differentiated features of Intel’s 300mm wafer silicon photonics platform is its integrated hybrid silicon laser. The new multiwavelength advancement enables the production of the optical source for future high-volume applications such as co-packaged optics or optical interconnects, facilitating the growing bandwidth requirements of data-intensive applications such as  artificial intelligence (AI) and machine learning (ML). 

More specifically, the laser uses a technology called dense wavelength division multiplexing (DWDM) in order to send different, closely spaced wavelengths over the same optical link. This technique thus increases the bandwidth while also reducing the physical size of the photonic chips. The key challenges that were overcome were the uniformity of the power and wavelength spacing. 

Intel’s current silicon photonics transceivers use coarse wavelength division multiplexing (CWDM) where the wavelengths are further apart. These products use 100G lines using four wavelengths that deliver 25G bandwidth each.   .... '