Ismail Ilkan Ceylan
Dr Ismail Ilkan Ceylan
Interests
My research interests are broadly in AI & machine learning with a particular focus on graph machine learning, which includes a class of challenging problems that can be naturally characterized using relational structures. The goal is to more efficiently and reliably learn from relational patterns and reason over them. This is a highly interactive field, where techniques from machine learning (e.g., deep learning, graph representation learning, probabilistic methods), knowledge representation (e.g., logical reasoning), and theoretical computer science (complexity theory, graph theory) are relevant. These methods are applied in a wide range of domains, ranging from systems in life-sciences (e.g., physical, chemical, and biological systems) to social networks.
Note: If you are interested in applying for a DPhil, it might be helpful to have a brief look at my graph representation learning course.
Selected Publications
-
Link prediction with relational hypergraphs
Xingyue Huang‚ Miguel Romero Orth‚ Pablo Barceló‚ Michael M Bronstein and İsmail İlkan Ceylan
In TMLR. 2025.
Details about Link prediction with relational hypergraphs | BibTeX data for Link prediction with relational hypergraphs
-
Equivariance Everywhere All At Once: A Recipe for Graph Foundation Models
Ben Finkelshtein‚ İsmail İlkan Ceylan‚ Michael Bronstein and Ron Levie
In Proceedings of the 39th Annual Conference on Neural Information Processing Systems (NeurIPS). 2025.
Details about Equivariance Everywhere All At Once: A Recipe for Graph Foundation Models | BibTeX data for Equivariance Everywhere All At Once: A Recipe for Graph Foundation Models
-
Curly Flow Matching for Learning Non−gradient Field Dynamics
Katarina Petrović‚ Lazar Atanackovic‚ Viggo Moro‚ Kacper Kapuśniak‚ İsmail İlkan Ceylan‚ Michael Bronstein‚ Michael Bronstein‚ Avishek Joey Bose and Alexander Tong
In Proceedings of the 39th Annual Conference on Neural Information Processing Systems (NeurIPS). 2025.
Details about Curly Flow Matching for Learning Non−gradient Field Dynamics | BibTeX data for Curly Flow Matching for Learning Non−gradient Field Dynamics