Data Science Investment Counter —Funding raised by UK data science companies in 2018.
£ 5.640 Million

Machine learning set to boost the efficiency of optical networks

Visualization from the Opte Project of the various routes through a portion of the Internet
Visualisation from the Opte Project of the various routes through a portion of the Internet

Machine learning could increase the efficiency of optical telecommunications networks (OTNs), new research has revealed.

Together with tech-giant Huawei, a group of researchers from Universitat Politècnica de Catalunya in Barcelona have retooled an artificial intelligence technique used for chess and self-driving cars to make OTNs run more efficiently. 

The results of the research, which will be presented at the Optical Fiber Conference and Exposition in San Diego next week, aim at reducing the buffering problems of traditional OTNs.

Since OTNs require rules to optimise the high amounts of traffic they manage, what this technology need is a better traffic guard.

According to the scientists who worked on the research, the new approach combines two machine learning techniques to solve this problem. The first, called reinforcement learning, creates a virtual “agent” that learns through trial and error the particulars of a system to optimise how resources are managed.

The second, called deep learning, adds an extra layer of sophistication to the reinforcement-based approach by using so-called neural networks – computer learning systems inspired by the human brain – to draw more abstract conclusions from each round of trial and error.

“Deep reinforcement learning has been successfully applied to many fields,” said one of the researchers, Albert Cabellos-Aparicio. “However, its application to computer networks is very recent. We hope that our paper helps kickstart deep-reinforcement learning in networking and that other researchers propose different and even better approaches.”

And in fact, if deep reinforced learning has been used before to optimise some resource allocation in OTNs, it has proven unable to process new scenarios.

The element that distinguishes Cabellos-Aparicio team’s research was how easily the new approach was able to learn about the networks after starting out with a blank slate.

“This means that without prior knowledge, a deep reinforcement learning agent can learn how to optimize a network autonomously,” Cabellos-Aparicio said. “This results in optimisation strategies that outperform expert algorithms.”

Thinking about future developments, the group plans to apply their deep reinforcement approach in combination with graph networks, an emerging field within artificial intelligence with the potential to transform scientific and industrial fields, such as computer networks, chemistry and logistics.

Image via Opte Project


Co-working space and blog dedicated to all things data science.

Subscribe to our newsletter