The Victor Albeck Award: He makes artificial intelligence even smarter

You encounter it every day. Whether using ChatGPT, signing in to online banking, entering a climate-controlled room, checking Facebook or being scanned in a hospital. In a short space of time, artificial intelligence has become an integral part of our everyday lives and offers fast and efficient solutions to a wide range of problems.

But it can be even better.

At the Department of Electrical and Computer Engineering, Professor Alexandros Iosifidis and his team are revolutionising AI. Immersed in data, algorithms and humming supercomputers, they are working to make artificial intelligence even more – yes, intelligent. One of the ways they are doing this is by creating new types of neural networks – a special type of artificial intelligence inspired by the structures of the human brain.

“Machine Learning and AI are growing very rapidly, and automatic, intelligent systems will play a far greater role in society and our daily lives in the future. We are working to solve machine learning problems far more efficiently and reliably and with much less computing power than is possible today. Our research enables machines to learn even better and perform tasks even faster. The goal is greater automation within a wide range of application areas,” says Alexandros Iosifidis.

Robots, drones and intelligent buildings

The Greek-born Professor's research has already been used in a wide range of contexts. He constantly pushes the boundaries of what is possible. He has worked to support older people to live independently in their homes using robotics, he has developed algorithms for the financial sector that can prevent stock market fraud, and he has improved the ability of drones to analyse visual information from the air.

Alexandros Iosifidis is currently involved in several projects, including the European project PANDORA. In this research initiative, he collaborates with a range of partners to create AI solutions that can make buildings more intelligent. The aim of the project is to make the indoor climate more adaptable to the users’ needs and more comfortable in relation to temperature changes:

“In an intelligent building, a network of intelligent IoT sensors helps to keep the indoor climate comfortable. In a centralised paradigm for artificial intelligence, the information from these sensors must be transmitted to a server that performs the analyses. But we are trying to decentralise AI so that the calculations take place on the data sources – e.g. the sensors – so that we avoid transferring data. In this way, AI becomes faster and more efficient, and we save power and avoid transferring personal information,” says Alexandros Iosifidis.

Democratisation of AI

By reducing the calculations and making AI even more efficient, the technology becomes accessible to people without specialised equipment. According to Alexandros Iosifidis, democratising artificial intelligence in this way is an important goal:

“Today, it is mainly large companies with large data centres that can carry out AI processes. This means that smaller companies that do not have access to these facilities lag behind. Democratising AI means that everyone can use technology to improve their operations, their business and their daily life. This is also in line with the EU's new AI Act on the responsible and sustainable use of AI,” says Alexandros Iosifidis.

Some people are concerned about the potential negative implications of artificial intelligence. Is the researcher himself concerned?

“Yes, of course. But it is important to understand that AI is a tool. A tool is neither positive nor negative in itself but can be used in a positive or negative way. You can use a knife to kill a person, but we all have knives in our houses,” says Alexandros Iosifidis.

So how do we make sure we're using AI in the right way? Alexandros Iosifidis believes we need legislation for the responsible use of AI – legislation that is already well underway.

The EU Parliament has already come up with a number of laws for the use of AI, and other countries will slowly start to follow suit. We are still in the early stages of this development, but I am positive that we will end up using AI responsibly. Above all, we must remember that AI can help people in so many different ways,” says Alexandros Iosifidis, who is proud to receive the Victor Albeck Award:

“It is a great recognition, not only for me but also for our department and our research team. I would really like to highlight the great progress that the Department of Electrical and Computer Engineering has made. The department is relatively new, but in a short space of time we’ve reached international standards within research and technology. And these standards are very high.”