Evolutionary Intelligence

Evolutionary Intelligence refers to self-learning systems that are biologically inspired. It can be used to solve complex real-world problems that are either difficult or impossible to solve optimally with exact algorithms. The algorithms and models that we examine are based on natural and biological systems, which are known to achieve extraordinary feats. Key examples are natural evolution and biological brains, the artificial equivalents of Evolutionary Algorithms (EAs) and Neural Networks (NNs).

The EI group's primary objective is to innovate such algorithms and systems by realizing novel synergies between Artificial Intelligence (AI) techniques, including EAs and NNs. A key example is our long-standing research line on the Gene-pool Optimal Mixing Evolutionary Algorithm (GOMEA) family.
Another major objective of the EI group is to create societal impact through our research. To achieve this, we collaborate with both industrial and societal partners. EI has a wide range of applications, leading to an expanding list of collaborations across various sectors, such as healthcare, logistics, agriculture, and industrial design.
Currently, most of our applications are in the field of medicine. Like the automated design of inherently explainable predictive systems that can link certain patient characteristics to the outcome of a treatment. Also, we continue our previously award-winning collaboration with Amsterdam UMC and Leiden UMC to tailor EI technology to optimize brachytherapy treatment plans. Brachytherapy is an essential type of radiotherapy for cancer treatment, where radioactive sources are placed in or near the tumor.

• More information on the Evolutionary Intelligence research group and its vision.
• More information on our Evolutionary Intelligence seminar.