Janis Kalofolias
Im Oberem Werk 1
66386 St. Ingbert, Germany
janis.kalofolias@cispa.de
Janis Kalofolias is a post-doctoral researcher affiliated with the EDA group. His research interests include many things, ranging from optimistic estimators for subgroup discovery, kernel-based methods for measuring similarities between graphs, to information theoretic methods for subjectively interesting structure from complex data.
Janis obtained his Bachelor of Science in 2011 from the University of Patras, Greece. In 2012 he joined Saarland University to pursue a Master of Science in Computer Science, and was a Research Assistant at the Max Planck Institute for Informatics. He joined the EDA group as a PhD student in November 2016, and defended his dissertation titled 'Subgroup Discovery for Structure Targets' on December 8th 2022. He subsequently was a Postdoc in the group until June 2024.
Publications
2024 | |
 |
Learning Exceptional Subgroups by End-to-End Maximizing KL-divergence. In: Proceedings of the International Conference on Machine Learning (ICML), PMLR, 2024. (spotlight, 3.5% acceptance rate; 27.5% overall)  |
2022 | |
|
Omen: Discovering Sequential Patterns with Reliable Prediction Delays. Knowledge and Information Systems vol.64(4), pp 1013-1045, Springer, 2022. (IF 2.822)  |
|
Naming the most anomalous cluster in Hilbert Space for structures with attribute information. In: Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), AAAI, 2022. (15.0% acceptance rate) |
|
|
2021 | |
|
SUSAN: The Structural Similarity Random Walk Kernel. In: Proceedings of the SIAM International Conference on Data Mining (SDM), SIAM, 2021. (21.2% acceptance rate) |
2019 | |
|
Discovering Robustly Connected Subgraphs with Simple Descriptions. In: Proceedings of the IEEE International Conference on Data Mining (ICDM), IEEE, 2019. (18.5% acceptance rate) |
|
Discovering Robustly Connected Subgraphs with Simple Descriptions. In: Proceedings of the ECMLPKDD Workshop on Graph Embedding and Mining (GEM), 2019. (oral presentation, 21% acceptance rate) |
|
Discovering Robustly Connected Subgraphs with Simple Descriptions. In: Proceedings of the ACM SIGKDD Workshop on Mining and Learning from Graphs (MLG), 2019. |
2017 | |
|
Efficiently Discovering Locally Exceptional yet Globally Representative Subgroups. In: Proceedings of the IEEE International Conference on Data Mining (ICDM'17), IEEE, 2017. (full paper, 9.3% acceptance rate; overall 19.9%) |
2016 | |
|
From Sets of Good Redescriptions to Good Sets of Redescriptions. In: Proceedings of the IEEE International Conference on Data Mining (ICDM), IEEE, 2016. (full paper, 8.5% acceptance rate; overall 19.6%) (invited for the KAIS Special Issue on the Best of IEEE ICDM 2016) |