The German multinational engineering and electronics company Bosch organized for the first time the Bosch AI Young Research award, endowed with 50,000 euros, the first edition of which was awarded to Gergely Neu, a researcher of the Artificial Intelligence and Machine Learning (AI&ML) research group at the Department of Information and Communication Technologies (DTIC) at UPF, in recognition of the work he is carrying out. The award was given on 29 October at the Bosch AI CON 2019 international conference in Renningen (Germany).

As the award panel stated, “he has been granted this award in recognition of his research in basic reinforcement learning”, an area that belongs to the field of artificial intelligence (AI), which has multiple applications for industry and is mainly devoted to finding algorithms capable of making decisions among an infinite number of situations that can be combined in numerous ways.

Reinforcement learning: when machines learn as children do

For over ten years, the Hungarian scientist has performed his research in the field of reinforcement learning (RL). RL is an area of machine learning in which algorithms intuitively learn and experiment how their environment is constructed and what rules apply, in the same way that a child would explore the world and discover how it works.

Neu exposes algorithms to new environments and unfamiliar situations, asks them to take a large number of decisions, and then examines the factors that have led to success or failure

In artificial intelligence applications, such as highly automated vehicles, automated financial trading systems or smart grids, reliability and stable performance are essential. Neu investigates at what time RL algorithms reach their limits and why. He then uses the results to develop increasingly robust algorithms that work more reliably.

With his team, Neu develops algorithms under as real conditions as possible: he exposes algorithms to new environments and unfamiliar situations, asks them to take a large number of decisions, and then examines the factors that have led to success or failure.

Uniting the theory and practice of artificial intelligence

Gergely Neu considers two-way exchange between academia and industry as being essential, as mutual benefit is the only way for each area to be able to expand its knowledge. “I am deeply honoured to accept this award and I am especially pleased to see Bosch’s commitment to supporting academic research”, Neu stated. “In recent years, many talented researchers in artificial intelligence have left academia for lucrative jobs in industry, so prizes like the Bosch AI Young Research Award play an important role in increasing the prestige of traditional academic careers”, he continued.

Neu plans to invest the 50,000 euro prize to expand his group’s current collaborations and also to create new ones by inviting researchers to his laboratory and allowing his own team to visit researchers at other labs and attend conferences.

Gergely Neu joined Pompeu Fabra University in 2015. He holds a PhD from Budapest University of Technology and Economics (2013) and was a postdoctoral researcher at the Research Institute for Digital Sciences, INRIA, in Lille (France). His most recent appointment was as a visiting researcher to Google Brain in Zurich (Switzerland), where he focused on the relevance of science for industry, which he plans to finish developing in 2020 at the Simons Institute of the University of California, Berkeley (USA).

He has recently been recognized for his career track record on several occasions. In February 2018 he obtained a “la Caixa” Junior Leader Retaining Fellowship; in March 2019 won the Google Faculty Research Award for his contributions to reinforcement learning methods.