George Dasoulas

George Dasoulas

2022 Wojcicki Troper HDSI Postdoctoral Fellow
Department of Biomedical Informatics
Harvard Medical School
George Dasoulas

George sits within Zitnik Lab at the Department of Biomedical Informatics of Harvard University where he concentrates in learning graph representations for biomedical applications.

George Dasoulas works on machine learning for structured data with a particular emphasis on graphs and generally non-Euclidean structures. His research focuses on building powerful graph learning models, that are able to extract knowledge in diverse real-world applications, ranging from communication networks to bioinformatics. During his research journey so far, George encountered challenging problems in the expressivity of Graph Neural Networks for molecular structure modeling, contributing to more accurate predictions in bioinformatics tasks. In parallel, he met the demanding needs of the telecommunications industry, where graph learning models have shown a crucial impact. As an HDSI fellow, he is interested in leveraging the computational capabilities of graph learning models for accelerating biomedical research.

George received a Ph.D. in Computer Science from Laboratoire d'informatique at École Polytechnique in Paris, France under the supervision of Prof. Michalis Vazirgiannis. During his doctoral studies, he worked as a doctoral researcher in the telecommunications industry. Before that, he received a Diploma in Electrical and Computer Engineering from the National Technical University of Athens, Greece.