Gergely Neu

Gergely Neu is a Research Assistant Professor in the Department of Information and Communication Technologies (DTIC) at Universitat Pompeu Fabra in Barcelona, Spain. He leads a research group focused on theoretical aspects of sequential decision making and machine learning.

Dr. Neu's research primarily centers on theoretical machine learning with a focus on:

• Reinforcement Learning Theory
• Online Optimization
• Bandit Problems
• Statistical Learning Theory
• Generalization Bounds
• Algorithms with provable performance guarantees

His work emphasizes developing algorithms with strong theoretical guarantees in both computational and statistical complexity while maintaining practical implementability. Dr. Neu's research program seeks to bridge the gap between theoretical foundations and practical applications of machine learning systems, particularly in sequential decision-making contexts where performance guarantees are crucial.

Analysis of Dr. Neu's recent publications reveals a strong focus on advancing the theoretical understanding of reinforcement learning, bandit algorithms, and generalization in machine learning. His work often employs techniques from convex analysis, information theory, and regret analysis to derive novel theoretical results with practical implications. A notable trend in his recent work is the exploration of offline reinforcement learning, contextual bandits, and the development of tighter generalization bounds using innovative analytical approaches that connect online learning with PAC-style guarantees.

Dr. Neu has received numerous prestigious awards and recognitions for his research contributions:

• ERC Starting Grant (2020)
• Bosch AI Young Researcher Award (2020)
• Google Faculty Research Award (2019)
• "la Caixa" Junior Leader fellowship (2018)
• ELLIS Scholar appointment (2021)
• Outstanding Reviewer Award from ICML'16
• Blue Ribbon for Outstanding Reviewing from IJCAI'16

Dr. Neu actively mentors several PhD students and postdoctoral researchers, fostering the next generation of theoretical machine learning researchers. His ERC Starting Grant provides substantial funding for his research program, enabling him to pursue ambitious theoretical investigations while training early-career researchers. He has successfully placed alumni in prestigious positions including assistant professorships at TU Delft and Politecnico di Milano.

Dr. Neu's research group operates within the vibrant AI and Machine Learning research community at Universitat Pompeu Fabra, collaborating closely with other researchers in the Department of Information and Communication Technologies and maintaining strong ties with the Statistics group at the Department of Economics and Business. The group participates actively in the international machine learning theory community, regularly contributing to top conferences such as COLT, NeurIPS, ICML, and ALT, where Dr. Neu has served in leadership roles including Program Chair for COLT'23.