Gal Elidan

Gal Elidan is an associate professor in the Department of Statistics at the Hebrew University of Jerusalem, Faculty of Social Sciences. His work bridges statistical theory with practical applications in machine learning and artificial intelligence.

Dr. Elidan's research focuses on Probabilistic Graphical Models, with emphasis on fundamental representations and methods for inference and learning in large-scale domains. His work addresses structure learning, discovery of hidden variables, transfer of knowledge between related tasks, and nonlinear high-dimensional representations of continuous or hybrid distributions. He also applies these techniques to real-life applications including computational biology, machine vision, and medical informatics, with recent focus on principled techniques for medical diagnosis.

His publication record shows consistent contributions to top machine learning venues with a recent emphasis on copula models for high-dimensional data. His work demonstrates a progression from foundational graphical model research to specialized applications in medical informatics and time series analysis.

Scientific recognition includes:

• Runner-up for Best Student Paper Award at UAI 2004
• Best Paper Award at ISMB 2001

Dr. Elidan serves on the editorial board of the Journal of Artificial Intelligence Research (JAIR) and has contributed to the development of important software tools including FastInf, a library for large-scale inference of graphical models, and LibB, a package for Bayesian network inference and learning. He has organized significant academic events including the NIPS 2011 Workshop on Copulas in Machine Learning and co-organized the PASCAL2 Probabilistic Inference Challenge and UAI 2010 Approximate Inference Challenge.