Johannes Lutzeyer -- Personal Website

I am Johannes Lutzeyer an Assistant Professor in the Data Science and Mining Team at the Laboratoire d'Informatique of École Polytechnique in France. Previously, I completed a 2.5 year postdoc, under the supervision of Prof. Michalis Vazirgiannis, at École Polytechnique and a PhD thesis on the spectral properties of the adjacency and Laplacian matrices under the supervision of Prof. Andrew Walden at Imperial College London.

My current research focuses on Graph Neural Networks and Spectral Properties of Graph Shift Operator Matrices. As such, I work in the area of Graph Representation Learning in the intersection of Statistics and Computer Science.

My academic CV can be accessed here.

News

September 2025: Jesse Read and I invite applications for a Postdoctoral Researcher position in the intersection of Reinforcement Learning, Graph Neural Networks and Graph Signal Processing. Please consider applying! [Description of Position]
August 2025: My coauthors, that is Abderrahim Bendahi, Adrien Fradin, Benjamin Doerr, and I are very happy to have received a distinguished paper award at IJCAI 2025 for our work.
August 2025: I gave a seminar talk entitled “Recent Advances on Graph Neural Networks” at the VisualAI lab at Princeton upon invitation of Ye Zhu and Olga Russakovsky.
July 2025: I gave a keynote talk entitled “Avenues for Interaction between Bayesian Methodology and Graph Neural Networks” at the “Combining Bayesian and Neural approaches for Structured Data Workshop” at IJCNN organised by Federico Errica, Daniele Castellana, Marco Podda, Davide Bacciu and colleagues.
June 2025: I gave a talk entitled “Frustration Tolerance and Other Skills To Aquire” at the LIX PhD and PostDoc Seminar “Dreams vs Reality” upon invitation of Théo Boury, Bernardo Hummes Flores and Sarah Berkemer.
April 2025: I gave a talk entitled “We Need Metrics for the Localisation and Factorisation of Learning Tasks on Graphs” at the NITMB Workshop on “New Developments in the Theory and Methodology of Graph Neural Networks” upon invitation of Claire Donnat and Olga Klopp. [Slides]
March 2025: I gave a keynote talk entitled “Understanding Virtual Nodes in Graph Neural Networks: Oversmoothing, Oversquashing and Node Heterogeneity “ at the GdR Day on Learning and Graphs organised by Nicolas Keriven and Pierre-Henri Paris. [Slides]
March 2025: I gave a talk entitled “An Analysis of Virtual Nodes in Graph Neural Networks” at the GeomeriX group seminar upon invitation of Julie Mordacq, Jiong Chen and Steve Oudot.
January 2025: [Applications are now closed.] Jesse Read and I invite applications for a PhD position with planned starting date in September 2025. Please consider applying! [Description of Position]
December 2024: I gave a talk entitled “Methodological Advances in Graph Neural Networks” at Ericsson Research upon invitation of Anastasios Giovanidis.
November 2024: I am very happy to receive one of the LOG Top Reviewer Awards.
November 2024: Please come join us at the Paris Learning on Graphs Meetup accompanying the Learning on Graphs Conference.
August 2024: [Applications are now closed.] We invite applications for a CIFRE PhD position in collaboration with CMA-CGM on Graph Neural Networks for Maritime Trade.
July 2024: I am very happy to receive one of the ICML Best Reviewer Awards.
June 2024: I gave a keynote talk entitled “Graph Learning for and with Graph Neural Networks” at the GdR Graph Learning Day organised by Arnaud Breloy and Benjamin Girault. [Slides]
May 2024: I gave a seminar entitled “Recent Advances in Graph Neural Network Robustness” at the AIDRC Seminar Series organised by Mohamed El Amine Seddik. [Slides]
December 2023: I am very happy to receive one of the NeurIPS Top Reviewer Awards.
November 2023: Please come join us at the Paris Learning on Graphs Meetup accompanying the Learning on Graphs Conference.
July 2023: I gave a plenary talk entitled “Path Neural Networks: Expressive and Accurate Graph Neural Networks” at the 2nd ELLIS UnConference organised by Alain Durmus and colleagues.
July 2023: I gave a seminar entitled “Recent Advances in Graph Neural Networks” at the Graphs Guild of Lloyds Banking Group upon invitation of Jack Evan Rodwell.
June 2023: I gave a talk entitled “Message Passing In Graph Neural Networks” at the Learning on Graphs Meetup Paris organised by Alexandre Duval and Fragkiskos Malliaros. [Slides]

Publications

S. Ennadir, O. Smirnov, Y. Abbahaddou, L. Cao & J. F. Lutzeyer, “Enhancing Graph Classification Robustness with Singular Pooling,” Conference on Neural Information Processing Systems (NeurIPS), 2025.
Y. Abbahaddou, F. D. Malliaros, J. F. Lutzeyer, A. M. Aboussalah & M. Vazirgiannis, “ADMP-GNN: Adaptive Depth Message Passing GNN,” Conference on Information and Knowledge Management (CIKM), 2025.
Y. Abbahaddou, F. D. Malliaros, J. F. Lutzeyer, A. M. Aboussalah & M. Vazirgiannis, “Graph Neural Network Generalization with Gaussian Mixture Model Based Augmentation,” International Conference on Machine Learning (ICML), 2025.
A. Bendahi, B. Doerr, A. Fradin & J. F. Lutzeyer, “Speeding Up Hyper-Heuristics With Markov-Chain Operator Selection and the Only-Worsening Acceptance Operator,” International Joint Conference on Artificial Intelligence (IJCAI), 2025.
R. Zein-Eddine, M. Ramuz, G. Refrégier, J. F. Lutzeyer, A. Aleksandrov & H. Myllykallio, “Understanding the key challenges in tuberculosis drug discovery: what does the future hold?,” Expert Opinion on Drug Discovery, 2025.
N. Kormann, M. Ramuz, Z. Nisar, N. S. Schaadt, H. Annuth, B. Doerr, F. Feuerhake, T. Lampert & J. F. Lutzeyer, “HIEGNet: A Heterogenous Graph Neural Network Including the Immune Environment in Glomeruli Classification,” Medical Imaging with Deep Learning Conference (MIDL), 2025.
J. Southern, F. Di Giovanni, M. Bronstein & J. F. Lutzeyer, “Understanding Virtual Nodes: Oversquashing and Node Heterogeneity,” International Conference on Learning Representations (ICLR), 2025.
S. Ennadir, J. F. Lutzeyer, M. Vazirgiannis & E. H. Bergou, “If You Want to Be Robust, Be Wary of Initialization,” Conference on Neural Information Processing Systems (NeurIPS), 2024.
Y. Abbahaddou, S. Ennadir, J. F. Lutzeyer, M. Vazirgiannis & H. Boström, “Bounding the Expected Robustness of Graph Neural Networks Subject to Node Feature Attacks,” International Conference on Learning Representations (ICLR), 2024.
S. Ennadir, Y. Abbahaddou, J. F. Lutzeyer, M. Vazirgiannis & H. Boström, “A Simple and Yet Fairly Effective Defense for Graph Neural Networks,” AAAI Conference on Artificial Intelligence (AAAI), 2024.
Y. Abbahaddou, J. F. Lutzeyer & M. Vazirgiannis, “Graph Neural Networks on Discriminative Graphs of Words,” NeurIPS New Frontiers in Graph Learning Workshop, 2023.
G. Michel, G. Nikolentzos, J. F. Lutzeyer & M. Vazirgiannis, “Path Neural Networks: Expressive and Accurate Graph Neural Networks,” Proceedings of the 40th International Conference on Machine Learning (ICML), 2023.
B. Doerr, A. Dremaux, J. F. Lutzeyer & A. Stumpf, “How the move acceptance hyper-heuristic copes with local optima: drastic differences between jumps and cliffs,” Proceedings of the Genetic and Evolutionary Computation Conference (GECCO), 2023.
M. Chatzianastasis, J. F. Lutzeyer, G. Dasoulas & M. Vazirgiannis, “Graph Ordering Attention Networks,” AAAI Conference on Artificial Intelligence (AAAI), 2023.
G. Salha-Galvan, J. F. Lutzeyer, G. Dasoulas, R. Hennequin & M. Vazirgiannis, “New Frontiers in Graph Autoencoders: Joint Community Detection and Link Prediction,” NeurIPS New Frontiers in Graph Learning Workshop, 2022.
A. R. Ramos Vela, J. F. Lutzeyer, A. Giovanidis & M. Vazirgiannis, “Improving Graph Neural Networks at Scale: Combining Approximate PageRank and CoreRank,” NeurIPS New Frontiers in Graph Learning Workshop, 2022.
A. Qabel, S. Ennadir, G. Nikolentzos, J. F. Lutzeyer, M. Chatzianastasis, H. Bostrom & M. Vazirgiannis, “Structure-Aware Antibiotic Resistance Classification Using Graph Neural Networks,” NeurIPS AI for Science Workshop, 2022.
G. Salha-Galvan, J. F. Lutzeyer, G. Dasoulas, R. Hennequin & M. Vazirgiannis, “Modularity-Aware Graph Autoencoders for Joint Community Detection and Link Prediction,” Neural Networks, vol. 153, pp. 474–495, 2022.
J. F. Lutzeyer*, C. Wu* & M. Vazirgiannis, “Sparsifying the Update Step in Graph Neural Networks,” ICLR Workshop on Geometrical and Topological Representation Learning, 2022.
M. E. A. Seddik, C. Wu, J. F. Lutzeyer & M. Vazirgiannis, “Node Feature Kernels Increase Graph Convolutional Network Robustness,” International Conference on Artificial Intelligence and Statistics (AISTATS), 2022.
G. Dasoulas*, J. F. Lutzeyer* & M. Vazirgiannis, “Learning Parametrised Graph Shift Operators,” International Conference of Learning Representations (ICLR), 2021.
J. F. Lutzeyer & A. T. Walden, “Comparing Spectra of Graph Shift Operator Matrices,” International Conference on Complex Networks and their Applications, 2020.
J. F. Lutzeyer & A. T. Walden, “Extending the Davis-Kahan theorem for comparing eigenvectors of two symmetric matrices I: Theory,” arXiv:1908.03462, 2019.
J. F. Lutzeyer & A. T. Walden, “Extending the Davis-Kahan theorem for comparing eigenvectors of two symmetric matrices II: Computation and Applications,” arXiv:1908.03465, 2019.
J. F. Lutzeyer & E. A. K. Cohen, “Correcting the estimator for the mean vectors in a multivariate errors-in-variables regression model,” arXiv:1510.03600, 2015.

* is used to denote equal contribution.

I would like to sincerely thank Guillaume Salha-Galvan for creating this website and allowing me to copy the format of his website. In turn, Guillaume would like to acknowledge the github pages theme by orderedlist.