This article was reviewed based on Science X’s editorial process and policies. The editors have highlighted the following attributes ensuring the credibility of the content:

verified

peer-reviewed publication

trusted source

correct






Memristive optical platforms for non-volatile transmission modulation. Credit: Photonics of nature (2023). DOI: 10.1038/s41566-023-01217-w

AI, machine learning, and ChatGPT may be relatively new buzzwords in the public domain, but developing a computer that functions like the human brain and nervous system, hardware and software combined, has been a decades-long challenge. University of Pittsburgh engineers are today exploring how optical ‘memristors’ may be a key to the development of neuromorphic computing.

Memory resistors, or memristors, have already demonstrated their versatility in electronics, with applications such as computational circuit elements in neuromorphic computing and compact memory elements in high-density data storage. Their unique design paved the way for in-memory computing and captured considerable interest from scientists and engineers.

A new review article published in Photonics of naturetitled “Integrated Optical Memristor”, sheds light on the evolution of this technology and the work that still needs to be done for it to reach its full potential.

Led by Nathan Youngblood, an assistant professor of electrical and computer engineering at the University of Pittsburgh’s Swanson School of Engineering, the paper explores the potential of optical devices that are analogous to electronic memristors. This new class of device could play an important role in revolutionizing high-bandwidth neuromorphic computing, machine learning hardware, and artificial intelligence in the optical domain.

“Researchers are truly intrigued by optical memristors because of their incredible potential in high-bandwidth neuromorphic computing, machine learning hardware, and artificial intelligence,” Youngblood explained. “Imagine merging the incredible benefits of optics with local information processing. It’s like opening the door to a whole new realm of technological possibilities that were previously unimaginable.”

The review article presents a comprehensive overview of recent advances in this emerging field of photonic integrated circuits. Explore the current state of the art and highlight potential applications of optical memristors, which combine the advantages of ultra-fast, high-bandwidth optical communication with local information processing. However, scalability has emerged as the most pressing issue that future research should address.

“Scaling in-memory or neuromorphic computation in the optical domain is a huge challenge. Having a technology that is fast, compact and efficient makes scaling more achievable and would be a huge step forward,” Youngblood explained.

“An example of the limitations is that if you were to take phase change materials, which currently have the highest storage density for optical memory, and try to implement a relatively simplistic on-chip neural network, it would take a wafer the size of a laptop to fit as many memory cells as needed,” he continued. “Size matters for photonics, and we need to find a way to improve storage density, power efficiency, and programming speed to perform useful calculations at useful scales.”

Using light to revolutionize computing

Optical memristors can revolutionize computation and information processing in several applications. They can enable active trimming of photonic integrated circuits (PICs), allowing on-chip optical systems to be tuned and reprogrammed as needed without continuously consuming power. They also offer high-speed data storage and retrieval, promising to speed up processing, reduce power consumption, and enable parallel processing.

Optical memristors can even be used for artificial synapses and brain-inspired architectures. Dynamic memristors with nonvolatile memory and nonlinear output replicate the long-term plasticity of synapses in the brain and pave the way for integration-and-fire computing architectures.

Research to expand and improve optical memristor technology could unlock unprecedented possibilities for high-bandwidth neuromorphic computing, machine learning hardware and artificial intelligence.

“We’ve looked at many different technologies. The thing we’ve noticed is that we’re still a long way from the goal of an ideal optical memristor, something that’s compact, efficient, fast, and changes optical properties significantly,” Youngblood said. “We’re still looking for a material or device that actually meets all of these criteria in one technology to advance the field.”

More information:
Nathan Youngblood et al, Integrated Optical Memristors, Photonics of nature (2023). DOI: 10.1038/s41566-023-01217-w

About the magazine:
Photonics of nature

#Optical #memristors #review #Shedding #light #neuromorphic #computation

By admin

Leave a Reply

Your email address will not be published. Required fields are marked *