European cooperation for a safe human-machine interaction

European cooperation for a safe human-machine interaction
The European Network of Artificial Intelligence Excellence Centres (HumanE-AI-Net) including researchers from ELTE has started its operation. The main objective of the H2020 project is to make AI systems more human-centric.

Led by the Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI), the HumanE-AI-Net project comprises thirty-five institutions from twelve member states, bringing together European research centres, universities, and industrial actors working in the field of artificial intelligence. The goal is to harness the emergence of human-centred and socially responsible artificial intelligence adapted to European conditions and needs, which is at the forefront of global scientific and technological development.

Broadening the research-development base for the current technological breakthrough, the HumanE-AI-Net research network contributes to the emergence of robust, reliable AI systems capable of appropriate human–machine interactions in a complex social environment. Technologically, the project focuses on machine learning, reasoning and design, multimodal perception and modelling, computer vision, speech recognition, as well as human–machine cooperation. Putting the new technological results into a broader social and ethical context, the starting research collaboration places great emphasis on the concept of artificial intelligence that assists, supports, and interacts with humans.

An important line of its activity is directed at the reliability of artificial intelligence,

the explainability of decisions made by machines, building up user trust, and clarifying the legal and ethical issues associated with artificial intelligence. In addition, it aims to encourage the development of innovation ecosystems, the mobility of researchers, and the training of professionals.

The research to be carried out by Eötvös Loránd University is meant to support the work of doctors and health care workers in diagnosis, therapy, and follow-up, as well as to aid monitoring and active assistance using augmented intelligence. In the field of psychiatry, the planned areas of application involve primarily post-traumatic stress disorder, depression, and autism. In the field of radiology, it will be applied in the evaluation of results produced by medical imaging procedures. The machine processing of large volumes of data from image, speech, and video analyses, as well as machine suggestions may provide support to physicians in decision-making.

The contribution of ELTE is coordinated by the Department of Artificial Intelligence at the Faculty of Informatics with the contribution of the Institute of Physics and Mathematics at the Faculty of Science. Several other departments of the Faculty of Informatics are involved in the work due to the complex software requirements of the interaction and the medical quality control regulations. At the Faculty of Science, the Department of Physics of Complex Systems and the Artificial Intelligence Research Group of the Institute of Mathematics promote the project with their experience in basic and applied research activities.

The establishment and operation of the centres of excellence under the European research programme of Horizon 2020 will create an excellent opportunity for researchers at ELTE to contribute to the implementation of the EU’s artificial intelligence strategy in collaboration with top European universities and leading industrial actors.