Gergely Neu, member of the Artificial Intelligence and Machine Learning research group (AI&ML) of UPF’s Department of Information and Communication Technologies (DTIC) de la UPF, has been awarded an ERC Starting Grant to the sum of around 1.5 million euros to undertake the project, Provably Efficient Algorithms for Large-Scale Reinforcement Learning (ScaleR) and study reinforcement learning in artificial intelligence, over the next five years.

Reinforcement learning methods have gained significantly in popularity as the driving force  behind spectacular advances in artificial intelligence, enabling these technologies to be applied to problems of the real world, such as in the control of autonomous vehicles or intelligent energy networks.

"We hope the proposed research will have a significant impact, on both a theoretical and practical level"

As Neu says: "With this fresh impulse, we hope the proposed research will have a significant impact in reinforcement learning, on both a theoretical and practical level, giving the methods in question a practical application".

For more than ten years, Gergely Neu has been conducting research into reinforcement learning (RL). RL is an area of automatic learning in which algorithms learn and experience how their surroundings are constructed and what rules apply intuitively, in the same way that a child explores the world and discovers how it works.

In reinforcement learning, algorithms learn and experience how their surroundings are constructed and what rules apply intuitively

Gergely Neu joined Pompeu Fabra University in 2015. He received his doctorate at the Budapest University of Technology and Economics (2013) and was a postdoctoral researcher at the INRIA digital science research institute in Lille (France). He has recently been recognised on several occasions for his professional work. In February 2018, he was awarded La Caixa’s Junior Leader Retaining Fellowship; in March 2019, he won the Google Faculty Research Award for his contributions to reinforcement learning methods and, also in 2019, the German multinational engineering and electronics company Bosch awarded him the Bosch AI Young Research Award.

ERC Starting Grants for other researchers in the UPF group

Besides Gergely Neu, the following people have been awarded ERC grants in the same call:

Victoria Vanasco, researcher at the International Economics Research Centre (CREI) for the project: Information, Markets, and the Macroeconomy.

Arnau Busquests García, researcher at the Mar Medical Research Institute (IMIM) for the project: High Memory Beyond classical conditioning: Hippocampal circuits in higher-order memory processes.

Marc Suárez-Calvet, researcher from the BarcelonaBeta Brain Research Centre for the project: Identification of age-related Human Blood factors as a therapeutic target for Alzheimer's disease.

Rosa Fernández, researcher from IBE-CSIC for the project: Land animal evolution: genomic landmarks on the path to terrestrial life.

Daniel Richter, researcher from IBE-CSIC, for the project: The missing majority: understanding global ocean ecology by revealing the unknown biology of the most abundant marine microbial eukaryotes.

In total, in this ERC Starting Grant 2020 call, 436 researchers from 40 different countries will each receive funding up to the sum of 1.5 million euros, amounting to a total of 677 million euros. This money will assist scientists and academics in their pioneering research projects in all disciplines within the European Union’s Horizon 2020 research and innovation programme.