Academy of Finland offers wide support for AI research

At the end of last year, the Academy of Finland granted more than 13 million euros in programme-based funding to support research into artificial intelligence (AI). Just over 6 million euros was granted to AI-related research projects under the ICT 2023 programme and 7 million euros to research projects under the Academy Programme AIPSE (Novel Applications of Artificial Intelligence in Physical Sciences and Engineering Research).

The research, development and innovation programme ICT 2023 is jointly coordinated by the Academy of Finland and Business Finland (formerly Tekes). The aim of the programme is to further improve Finland’s scientific expertise in computer science and to promote the extensive application of ICT. The AIPSE Academy Programme, in turn, promotes the utilisation of AI in physical sciences and engineering research.

Different types of data-driven methods are continuing to gain in importance in research, administration and industry. The rapid development of various methods that utilise AI owes much to the advances that have been made in machine learning, pattern recognition, statistics, data mining and computational and software-based database technology. Other factors contributing to the rise of AI include the exponential growth of computing power. These new methods have broad application prospects in scientific research, too.

Biomedical machine learning, computational research into nanomaterials and prediction of crop levels

Among the ICT 2023 funding recipients is a consortium headed by Professor Tapio Pahikkala (University of Turku) that aims to develop novel mathematical techniques for machine learning. The consortium will use new optimisation algorithms to improve machine learning methods used in biomedicine.

Professor Aki Vehtari (Aalto University), Professor Antti Honkela (University of Helsinki) and Professor Arto Klami (University of Helsinki), in turn, received funding for a project that will develop methods for assessing the quality of variational approximations and improve their accuracy. To be feasible for use in scalable AI applications, such methods must be both computationally efficient and reliable. The novel techniques will be implemented as part of leading open-source programming systems, which will guarantee that the results are widely applicable in the data science industry and in academic research.

Among the AIPSE funding recipients is a research project headed by Research Professor Pirjo Peltonen-Sainio (Natural Resources Institute Finland) and FiDiPro Professor Hiroshi Mamitsuka (Aalto University) that aims to develop computational approaches to boost crop production. The project’s primary objective is to attain a high predictive performance for crop production. Another goal is to improve simulation quality by using machine learning, which allows using new types of data, such as satellite images, which cannot be used by conventional simulators.

Academy Professor Hannu Häkkinen (University of Jyväskylä) and Professor Tommi Kärkkäinen (University of Jyväskylä) were granted AIPSE funding for a project aiming at new scientific breakthroughs in computational research of nanomaterials. The results of the project will help in designing novel nanoparticles for use in applications such as biological imaging, catalysis and biological sensors.

• Learn more about the AIPSE programme
• Learn more about the ICT 2023 programme