Written by Noa Cecilie Sæther
– I try to blend the idea of quantum into the framework of bio-inspired algorithms. My aim is to make it available for everyone.
So says Kazi Shah Nawaz Ripon, who has been an associate professor at OsloMet since august 2023.
– Quantum computers are not expected to be on our desks very soon, so I would like to implement it into classical computers.
Ripon first moved to Norway in 2009 for his PhD in computer science at the University of Oslo, where he researched artificial intelligence, machine learning and bio-inspired techniques to optimize real-world large scale multi-objective combinatorial optimization problems.
Specifically, he developed hybrid bio-inspired algorithms, several artificial intelligence based heuristics and applied the hybrid algorithms to several real-words NP-hard problems in industries.
Ever since, Ripon has continued exploring the same field.
Traditional Methods Fail for Large-Scale Combinatorial Optimization
Many real-world problems — from logistics and scheduling to network design and resource allocation — fall under the umbrella of combinatorial optimization. These problems require finding the best solution from a vast number of possible combinations, and as the problem size increases, the number of options grows exponentially.
Take the Travelling Salesman Problem (TSP) as a classic example. It asks:
“Given a list of cities and distances between them, what is the shortest possible route that visits each city once and returns to the starting point?”
While this sounds simple, the number of possible routes grows factorially with the number of cities. For just 61 cities, the number of possible routes exceeds the number of atoms in the observable universe.
This kind of combinatorial explosion makes traditional, brute-force approaches impractical. Instead, such problems demand more advanced strategies — like heuristics, metaheuristics, or machine learning-based methods — that can intelligently explore the search space and find near-optimal solutions within a reasonable time.
These challenges are at the heart of modern optimization, where efficiency, scalability, and adaptability are key to solving complex, real-world problems.
In the real world, this way of thinking could for example be used in hospital scheduling.
– An important real-world example of combinatorial optimization is hospital surgery scheduling. Each surgery must be coordinated with available operating rooms, surgeons, anesthesiologists, nurses, and equipment — all within strict time constraints. This is a massive operation. When you are given a time for the operation, you are linked to several factors: with certain operating theaters, doctors and nurses. Then what happens if an operation is delayed? Another patient cannot be operated. Traditionally, this can’t be solved. Therefore, if one surgery is delayed, it can cascade through the schedule, affecting multiple patients and staff. The number of possible combinations becomes overwhelming, especially in large hospitals.
This is where traditional methods break down. The complexity is simply too high to handle manually or through brute-force calculations. Instead, hospitals need smart optimization techniques that can adapt, prioritize, and find efficient schedules — improving both resource use and patient outcomes.
Inspired by nature
Additionally, bio-inspired algorithms can be implemented in such problems. They are, as the name indicates, based on nature. Ripon explains that there could for example be a parallell between the human immune system and email spam detection.
That is because our immune system protects us from pathogens, by sorting them out. Inside the body, there are still helpful, “good” pathogens. In the same way, emails that are detected as “bad” are sorted into the spam folder. That means the email algorithm can differentiate between useful and harmful, Ripon explains.
Moreover, when ants detect food, they return to the nest while depositing a pheromone, which helps the next ant to detect the same food. If several ants find a food source in different places, the closest one will return to the nest first. Therefore, the next ones will follow in its steps.
– We take inspiration from nature – humans, ants, birds, and then we develop those ideas into algorithms, Ripon explains.
Detecting fake news
Recently, Ripon published a paper on detecting fake news in social media, based on an artificial immune system, using a negative selection algorithm.
This method is inspired by the the biological immune system, generating a set of detectors that can distinguish between “normal” and “abnormal” patterns in a set of data.
– The idea is to collect a list of fake news online, and then implement it in the real world. The algorithm will be able to detect the fake news, Ripon explains.
– Could do a lot in the medical field
Nowadays, he’s also trying to implement quantum algorithms with evolutionary algorithms.
– Usually, when you hear about quantum technology, you think you’ll need a quantum computer. But we’re not actually sure when we can get one in our hands. My idea is to use the quantum concept in a classical algorithm that can be applied to classical computers.
– Nobody actually knows how to use quantum in their everyday life. My dream scenario is for people to be able to use quantum computing like their laptop. That’s what’s missing in the research field.
Ripon would also like to apply quantum technology to the medical domain, for example in image segmentation or brain tumor detection. – I would like to implement techniques that can handle uncertain delays and update the hospital scheduling automatically. You could probably do a lot in the medical field.
