New Research by Caltech Assistant Professor Unlocks Photonic Computing Power with Artificial ‘Life’ – Pasadena Now

Credit: Nicolle R. Fuller, Sayo Studio

The relentless quest for faster, smaller computers that can do more has led manufacturers to design ever smaller transistors that are now packed into computer chips by the tens of billions.

And so far, this tactic has worked. Computers have never been more powerful than now. But there are limitations: Traditional silicon transistors can only get so small because of the difficulties in manufacturing devices that, in some cases, are only a few tens of atoms wide. In response, researchers have begun developing computer technologies, such as quantum computers, that don’t rely on silicon transistors.

Another avenue of research is photonic computing, which uses light instead of electricity, similar to how fiber-optic cables have replaced copper wires in computer networks. New research by Alireza Marandi of Caltech, an assistant professor of electrical engineering and applied physics, uses optical hardware to make cellular automata, a type of computer model consisting of a “world” (a gridded area) containing “cells” ( each square of the grid) that can live, die, reproduce, and evolve into multicellular creatures with their own unique behaviors. These automatons have been used to perform computational tasks and, according to Marandi, are ideal for photonic technologies.

“If you compare an optical fiber to a copper cable, you can transfer information much faster with an optical fiber,” says Marandi. “The big question is, can we use that informational capacity of light for computation instead of just communication? To answer this question, we are particularly interested in thinking about unconventional computing hardware architectures that are better suited to photonics than digital electronics.”

Cellular automata

To fully understand the hardware designed by Marandi’s group, it is important to understand what cellular automata are and how they work. Technically speaking, they’re computational models, but that term does little to help most people understand them. It is more useful to think of them as simulated cells that follow a very basic set of rules (each type of automaton has its own set of rules). Incredibly complex behaviors can emerge from these simple rules. One of the best-known cellular automata, called The Game of Life or Conway’s Game of Life, was developed by English mathematician John Conway in 1970. It has just four rules that apply to a grid of “cells” that can be dead or alive. These rules are:

1. Any live cell with fewer than two live neighbors dies, as per subpopulation.
2. Any live cell with more than three live neighbors dies, as from overcrowding.
3. Any live cell with two or three live neighbors survives the next generation.
4. Any dead cell with exactly three live neighbors will come to life, as if by reproduction.

A computer running the Game of Life repeatedly applies these rules to the world cells live in at regular intervals, with each interval considered a generation. Within a few generations, those simple rules cause the cells to organize themselves into complex shapes with evocative names like loaf of bread, beehive, toad and heavy spaceship.

A “toad” as it would appear in Conway’s Game of Life.
A “heavy spaceship” as it would appear in Conway’s Game of Life.
Credit: Maxgyisawesome/Wikimedia Commons

Basic, or “elementary,” cellular automata like The Game of Life appeal to researchers working in mathematics and computer theory, but they may also have practical applications. Some of the elementary cellular automata can be used for random number generation, physics simulations, and cryptography. Others are as computationally powerful as conventional computer architectures, at least in principle. In some ways, these task-oriented cellular automatons are similar to an ant colony in which the simple actions of individual ants combine to perform larger collective actions, such as digging tunnels or collecting food and returning it to the nest. More “advanced” cellular automata, which have more complicated rules (although still based on neighboring cells), can be used for practical computational tasks such as identifying objects in an image.

Marandi explains: ‘Although we are intrigued by the kind of complex behaviors we can simulate with relatively simple photonic hardware, we are very excited by the potential of more advanced photonic cellular automata for practical computational applications.’

Ideal for photonic computation

Marandi says cellular automata are suited to photonic computation for a couple of reasons. Because information processing is extremely local (remember that in cellular automata, cells interact only with their immediate neighbors), they eliminate the need for much of the hardware that makes photonic computation difficult: the various gates, switches, and devices that are otherwise required for moving and storing light-based information. And the high-bandwidth nature of photonic computing means that cellular automata can run incredibly fast. In traditional computing, cellular automata might be designed in a computer language, which is built on another “machine” language level below that, which in turn sits on top of the binary zeros and ones that make up digital information.

By contrast, in Marandi’s photonic computing device, the cells of the cellular automaton are just ultrashort pulses of light, which can enable operations up to three orders of magnitude faster than in the fastest digital computers. Because these pulses of light interact with each other in a hardware grid, they can process information on the go without being slowed down by all the layers that underpin traditional computing. In essence, traditional computers run digital simulations of cellular automata, but Marandi’s device runs actual cellular automata.

“The ultra-fast nature of photonic operations and the possibility of photonic cellular automata being built on chips could lead to next-generation computers that can perform important tasks much more efficiently than digital electronic computers,” says Marandi.

The paper describing the work, titled “Photonic Elementary Cellular Automata for Simulation of Complex Phenomena,” appears in the May 30 issue of the journal Light: Science & Applications. The lead author is Gordon HY Li (MS ’22), PhD candidate in applied physics; with co-authors Christian R. Leefmans, PhD candidate in applied physics; and James Williams, an electrical engineering graduate student.

Funding for the research was provided by the United States Office of Army Research, the Air Force Office of Scientific Research and the National Science Foundation.

Get our Pasadena newspaper delivered to your inbox. Free.

Get all the latest Pasadena news, 10+ new stories every day, 7 days a week at 7am

#Research #Caltech #Assistant #Professor #Unlocks #Photonic #Computing #Power #Artificial #Life #Pasadena
Image Source : www.pasadenanow.com

Leave a Comment