News

Our group will present 10 papers at NeurIPS 2024 (incl. 1 oral presentation and 2 spotlights). Congratulations to the entire team and all co-authors. More details on the papers will follow soon. Neural Pfaffians: Solving Many Many-Electron Schrödinger Equations (Oral) (Nicholas Gao, Stephan…

Our group presents two works at ICML. The first is an oral presentation focusing on AI for science: Expressivity and Generalization: Fragment-Biases for Molecular GNNs Tom Wollschläger*, Niklas Kemper*, Leon Hetzel, Johanna Sommer, Stephan Günnemann (oral presentation) The second work considers…

We are happy to present the following work at the ICLR conference: From Zero to Turbulence: Generative Modeling for 3D Flow Simulation Marten Lienen, David Lüdke, Jan Hansen-Palmus, Stephan Günnemann We also have one paper at the AI4DifferentialEquations in Science workshop: On Representing…

Our group will present four papers at this year's NeurIPS. The works cover graph neural networks, ML robustness/certification and TPPs. Links to the papers/preprints will follow soon! Yan Scholten, Jan Schuchardt, Aleksandar Bojchevski, Stephan Günnemann Hierarchical Randomized Smoothing …

Our group has five papers at the 2023 International Conference on Machine Learning (ICML). Congratulations to all co-authors. Simon Geisler, Yujia Li, Daniel Mankowitz, Ali Taylan Cemgil, Stephan Günnemann, Cosmin Paduraru Transformers Meet Directed Graphs While transformers have become vital…

Our group will present four papers at this year's ICLR (incl. one spotlight). Congratulations to all co-authors! Jan Schuchardt, Tom Wollschläger, Aleksandar Bojchevski, Stephan Günnemann Localized Randomized Smoothing for Collective Robustness Certification (selected for spotlight…

Our group will present four papers at this year's NeurIPS. The works cover graph neural networks and ML robustness/certification. Links to the papers/preprints will follow soon! Jan Schuchardt, Stephan Günnemann Invariance-Aware Randomized Smoothing Certificates Incorporating…

Our group has three papers at the 2022 International Conference on Machine Learning (ICML). Congratulations to all co-authors. John Rachwan, Daniel Zügner, Bertrand Charpentier, Simon Geisler, Morgane Ayle, Stephan Günnemann Winning the Lottery Ahead of Time: Efficient Early Network Pruning …

Our group has six papers accepted at the 2022 International Conference on Learning Representations (ICLR) -- three of these as spotlight papers. The topics cover various fields of graph learning (e.g. GNNs in the context of quantum mechanical calculations and for spatio-temporal forecasting; DAG…

As an introduction and overview of the current research in adversarial robustness of graph neural networks, I have written a comprehensive survey paper. I hope it will be useful for (young) researchers and practitioners that are interested in this field.  The survey has appeared as a book chapter…

I am happy to announce that our group has six papers accepted at NeurIPS 2021! With works on ML for graphs, TPPs and flows, as well as robustness of ML methods and uncertainty, all of our core research directions are represented. Congratulations to my PhD students for this great success.  Simon…

Our Machine Learning seminar has received the teaching award of the department of computer science's student council (TeachInf Award) for the best elective course in the summer term 2020. We are very happy about this award and thank all students for the big interest in our courses!

Our group has three papers accepted at the 2021 International Conference on Machine Learning (ICML): Johannes Klicpera, Marten Lienen, Stephan Günnemann Scalable Optimal Transport in High Dimensions for Graph Distances, Embedding Alignment, and More International Conference on Machine…

Our group has two papers accepted at the 2021 International Conference on Learning Representations (ICLR). Jan Schuchardt, Aleksandar Bojchevski, Johannes Klicpera, Stephan Günnemann Collective Robustness Certificates: Exploiting Interdependence in Graph Neural Networks International Conference…

Our group has four papers accepted at NeurIPS 2020, including one oral presentation (top 1% of submitted works)! The works cover the full range of our core research topics: machine learning for graphs, machine learning for temporal data, and reliability of ML methods, i.e. robustness & uncertainty. …

Our Machine Learning lecture has received the teaching award of the department of computer science's student council (TeachInf Award) for the best mandatory lecture in the study year 2019/2020. We are very happy about this award and thank all students for the big interest in our lecture!

Our work "Efficient Robustness Certificates for Discrete Data: Sparsity-Aware Randomized Smoothing for Graphs, Images and More" on certifiable robustness for ML models handling discrete data (e.g. Graph Neural Networks) has been accepted at the International Conference on Machine Learning (ICML),…

We have two papers at this year’s KDD: "Scaling Graph Neural Networks with Approximate PageRank": In collaboration with colleagues from Google Research we have developed a highly scalable GNN able to handle massive graphs in single-machine and distributed environments. The paper got accepted as…

I have been awarded a 2020 Google Faculty Research Award in the area of Machine Learning and Data Mining! I am one of only 17 researchers worldwide selected in this area, and the only one from the EU. I am looking forward for some exciting research on graph neural networks.

I got invited as a keynote speaker at the leading European Machine Learning conference (ECML-PKDD 2020; The European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases). Looking forward to interesting discussions! I will also give a keynote talk at the…

We have 3 papers accepted at ICLR 2020, including two spotlight presentations! Directional Message Passing for Molecular Graphs (spotlight) Johannes Klicpera, Janek Groß, Stephan Günnemann Intensity-Free Learning of Temporal Point Processes (spotlight) Oleksandr Shchur, Marin…

Stefan Weißenberger has received the Rohde & Schwarz Best Bachelor Award for his thesis "Generalized Diffusion for Learning on Graphs". His work has been a cornerstone of our publication "Diffusion Improves Graph Learning" that has been published at the world leading Machine Learning and AI…

We have 4 papers accepted at NeurIPS 2019, including one spotlight presentation! Uncertainty on Asynchronous Time Event Prediction (spotlight) Marin Biloš, Bertrand Charpentier, Stephan Günnemann Certifiable Robustness to Graph Perturbations Aleksandar Bojchevski, Stephan Günnemann Diffusion…

I am happy to present our research results on adversarial robustness of machine learning methods for graphs as a keynote speaker at multiple events Graph Neural Network Summit, Google AI, Zurich, December 2019 International Conference on Computer Vision (ICCV), Workshop on Scene Graph…

Acceptance rate for oral papers: 9% Overall acceptance rate: 14%

Our work "Adversarial Attacks on Node Embeddings" has been accepted at the International Conference on Machine Learning, ICML 2019.

Our work "GhostLink: Mining Latent Influence Networks for Influence-aware Item Recommendation", where we tackle the problem of learning latent influence networks (i.e. an instance of graph inference), has been accepted at the International World Wide Web Conference. Congratulations to my…

We have two papers accepted at the International Conference on Learning Representations (ICLR): "Adversarial Attacks on Graph Neural Networks via Meta Learning" and "Predict then Propagate: Graph Neural Networks meet Personalized PageRank". Congratulations to all co-authors!

Our group has won the Best Research Paper award at KDD 2018 for the work "Adversarial Attacks on Neural Networks for Graph Data". The paper studies for the first time adversarial attacks to graphs, specifically focusing on state-of-the-art graph convolutional networks. KDD is the flagship data…

Our work on implicit generative models for graphs "NetGAN: Generating Graphs via Random Walks" has been accepted as a long/oral paper at ICML 2018! Congratulations to my co-authors Aleksandar Bojchevski, Oleksandr Shchur, and Daniel Zügner!

Our paper "Adversarial Attacks on Neural Networks for Graph Data" has been accepted as an oral/long paper at KDD 2018! In our paper we study the novel problem of adversarial machine learning for graphs, specifically considering state-of-the-art node classification approaches such as (deep) graph…

Our paper "Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking" has been accepted at the International Conference on Learning Representations (ICLR).

Our papers "An LSTM approach to Patent Classification based on Fixed Hierarchy Vectors" and "Making Kernel Density Estimation Robust towards Missing Values in Highly Incomplete Multivariate Data without Imputation" have been accepted at the SIAM International Conference on Data Mining (SDM 2018).…

Our paper "Bayesian Robust Attributed Graph Clustering: Joint Learning of Partial Anomalies and Group Structure" has been accepted at AAAI 2018.

We are currently offering student assistant positions (Hiwi/Tutor) for our lecture on Machine Learning. More details are available here.

The German Computer Science Society (Gesellschaft für Informatik, GI) has awarded a Junior-Fellowship to Prof. Dr. Stephan Günnemann. The fellowship is designated to early-career scientists for their excellent contribution to the field of computer science.  Official press release

Our paper "Robust Spectral Clustering for Noisy Data" has been accepted at KDD 2017. Congratulations to my co-authors Aleksandar Bojchevski and Yves Matkovic!

Our paper "Detection and Prediction of Natural Hazards using Large-Scale Environmental Data" has been accepted at SSTD 2017. Congratulations to all co-authors!

Our paper "Automatic Algorithm Transformation for Efficient Multi-Snapshot Analytics on Temporal Graphs" has been accepted at VLDB 2017. Congratulations to all co-authors!

Our group has received a Microsoft Azure Research Award from Microsoft Research!

Our paper "The Power of Certainty: A Dirichlet-Multinomial Model for Belief Propagation" has been accepted at the SIAM International Conference on Data Mining (SDM 2017). Congratulations to all my co-authors!

We are currently offering student assistant positions (Hiwi/Tutor) for our lecture on Mining Massive Datasets. More details are available here.

Our paper "ZooBP: Belief Propagation for Heterogeneous Networks" has been accepted at the International Conference on Very Large Data Bases (VLDB 2017). Congratulations to all my co-authors! 

Stephan Günnemann will give a presentation at the Berlin Big Data Center (TU Berlin, DFKI) on November 28, 2016. The title of the talk is "Beyond Independence: Efficient Learning Techniques for Networks and Temporal Data".

Our paper “SQL- and Operator-centric Data Analytics in Relational Main-Memory Databases ” has been accepted at the International Conference on Extending Database Technology (EDBT 2017). Congratulations to all my co-authors!

Our paper “Hyperbolae Are No Hyperbole: Modelling Communities That Are Not Cliques” has been accepted as a full paper at the IEEE International Conference on Data Mining (ICDM 2016). Congratulations to my co-authors Saskia Metzler and Pauli Miettinen! Our paper “EdgeCentric: Anomaly Detection in…

Our paper “Continuous Experience-aware Language Model” has been accepted at the ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) as a full paper + oral presentation. The acceptance rate was 8.9%.

Stephan Günnemann serves as co-chair of the organization committee of the 2017 International Conference on Very Large Data Bases (VLDB).

Our paper “MiMAG: Mining Coherent Subgraphs in Multi-Layer Graphs with Edge Labels” has been accepted at the Knowledge and Information Systems journal.

Stephan Günnemann has been invited as a speaker at the annual meeting of the “GI-Beirat der Universitätsprofessoren” (GIBU 2016).