ECML PKDD 2026

The Lamarr Institute for Machine Learning and Artificial Intelligence is a gold sponsor at this year’s European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD). As Europe’s flagship machine learning and data mining conference, ECML PKDD brings together a global audience to explore the latest breakthroughs in AI, data science, and theoretical Machine Learning. Get to know the Lamarr Institute and find out how you can work with us by visiting our booth and connecting with our researchers on site.

Connect with Lamarr Researchers

Lamarr researchers contribute to ECML PKDD 2026 in a variety of ways by presenting their exciting research findings. They look forward to engaging discussions and meeting you on the following occasions:

Research Papers and Lamarr Presenters

“Grokking Is a Geometric Phase Transition: Archetypal Analysis Reveals Causal Structure in Transformer Representations”
Tue, Sep 8, 12:00–12:15 pm | Building 1, Via Claudio, Room II1 | Research Track, S1.05 – Large Language Models I
(Lorenz Sparrenberg)

“Interactive Pareto navigation for deep multi-task learning”
Tue, Sep 8, 12:20–12:35 pm | Building 1, Via Claudio, Room II2 | Research Track, S1.03 – Optimization, Bandits & Online Learning I
(Augustina Chidinma Amakor)

Tue, Sep 8, 4:00–6:00 pm | Building 1, Via Claudio, Room I1 | Research Track & Journal Track, S3.13 – Federated Learning
(Michael Kamp)

“PATIENT+N: Profiling Feature Importance for Focal Epileptic Seizure Onset Detection for New Patients”
Tue, Sep 8, 5:40–5:55 pm | Building 1, Via Claudio, Room I2 | Research Track & Journal Track, S3.12 – Interpretability & Explainability II
(Uttam Kumar, Elena Demidova)

“Robust Wearable Analytics under Signal Corruption and Subject Mismatch”
Tue, Sep 8 | Building 1, Via Claudio, Room T4 | PhD Forum
(Gabriela Brüll)

“Visual Exploration of Rule-Based Model Logic”
Wed, Sep 9, 10:30 am–5:00 pm | Building 1, Via Claudio, Hall | Nectar Track, Poster Session
(Natalia and Gennady Andrienko)

“Detecting Stable Cross-Impact Patterns in Bivariate Time Series”
Wed, Sep 9, 10:45–11:00 am | Building 1, Via Claudio, Room T3 | Nectar Track
(Natalia and Gennady Andrienko)

“Learning Generalized Hessian on Graphs for Continuous Graph Learning”
Wed, Sep 9, 5:45–6:00 pm | Building 1, Via Claudio, Room II3 | Research Track, S3.27 – Representation, Alignment & Generative Learning
(Amitoz Azad, Elena Demidova)

Lamarr Fellowship Projects

Learn more about the projects by our Lamarr Fellows:

“Geometric and Information Compression of Representations in Deep Learning”
Tue, Sep 8, 10:30–10:45 am | Building 1, Via Claudio, Room II3 | Research Track, S1.01 – Representation Learning
(Linara Adilova)

“ASTRA: Adaptive Structure-Aware Post-Hoc Alignment of Knowledge Graph Embeddings”
Wed, Sep 9, 3:15–3:30 pm | Building 1, Via Claudio, Room I3 | Research Track, S2.22 – Neuro-Symbolic Learning
(Duygu Ekinci Birol)

Featured Workshop: QCDS 2026

Among the highlights of this year’s workshop program is the workshop on Quantification and Classification under Dataset Shift (QCDS 2026) on Monday, September 7, 2026, co-organized by Lamarr research associate Mirko Bunse. As a follow-up to the Learning to Quantify (LQ) workshop series, QCDS 2026 expands its focus to examine how classification and quantification interact when data distributions shift in real-world applications. The workshop brings together researchers and practitioners to discuss robust machine learning methods and emerging applications. Selected papers from the workshop will also have the opportunity to be published in a special issue of the Data Mining and Knowledge Discovery journal (Springer).

Workshop Contributions

“Explaining the Unsupervised: Human-in-the-Loop Supervision of Iterative Data Grouping”
Mon, Sep 7, 3:00–3:45 pm | Building 1, Via Claudio, Room T3 | XKDD Workshop
(Natalia and Gennady Andrienko)

“ATWL — A Formal Language for Representing, Comparing, and Reusing Visual Analytics Workflows”
Mon, Sep 7, 3:05–4:00 pm | Building 1, Via Claudio, Room I5 | HLDM Workshop
(Natalia and Gennady Andrienko)

“On the Edit Path to GNN Decisions”
Fri, Sep 11, 2:50–3:10 pm | Building 1, Via Claudio, Room T1 | MLG Workshop
(Florian Seiffarth)

Ann-Kathrin Oster

Head of Networking to the profile