Academic interests
Dr. Cini’s research focuses on machine learning methods for processing data on structured domains. In particular, he is interested in relational representations for time series analysis and processing graph-structured data. His work has found practical applications in energy analytics (load and PV production forecasting) and healthcare (biomedical signal processing). His recent research interests include uncertainty quantification for structured data and relational inductive biases in reinforcement learning.
Background
Dr. Andrea Cini is a SNSF postdoc fellow at the EPFL, working in the IMOS lab led by Prof. Olga Fink. He is also affiliated with IDSIA, at Università della Svizzera italiana (USI), and the Northernmost Graph Machine Learning Group, at UiT the Arctic University of Norway. Dr. Cini is a member of the ELLIS Society.
Before EPFL, he was an SNSF postdoc at the University of Oxford with Prof. Michael Bronstein and obtained his PhD from USI under the supervision of Prof. Cesare Alippi. In 2023, he was a visiting researcher at Imperial College London, working with Prof. Danilo Mandic. His research has been published in top-tier machine learning venues, including JMLR, NeurIPS, ICML, and ICLR. His work on graph-based forecasting has received the Informatics Europe Best Dissertation Award 2025. He is one of the creators of Torch Spatiotemporal.
Andrea Cini is affiliated with Integreat in the scope of onging collaborations with Associate Professor Filippo Maria Bianchi at UiT.