Joschka Groß

Saarland Informatics Campus

Campus E2 1 (CBI), Room 402

66123 Saarbrücken

Germany

About Me

I joined the Modeling and Simulation Chair in 2021 after completing my Master’s thesis on Graph Neural Networks (GNNs) for combinatorial optimization. I then worked at the intersection of GNNs and deep reinforcement learning for a year before joining the neuro-explicit drug discovery project.

Since then, I have been working on molecular machine learning tasks, with a focus on 3D-based geometric GNNs applied to kinase-focused drug discovery problems. My research explores how to best leverage the available biological data for training geometric GNNs and investigates which biophysical mechanisms these models are truly capable of capturing—using explainability methods and model-level interpretation techniques.

Interests

  • Molecular Machine Learning
  • Geometric Graph Neural Networks
  • Drug Discovery
  • Explainable ML

Education

Publications (10)

2026

3 papers

Assay-Based Machine Learning: Rethinking Evaluation in Drug Discovery

2026 — Digital Discovery

@article{grossAssayBasedMachineLearningRethinking2026,
 author = {Groß, Joschka and Backenköhler, Michael and Volkamer, Andrea},
 doi = {10.1039/D6DD00472E},
 eprint = {https://pubs.rsc.org/dd/article-pdf/doi/10.1039/D6DD00472E/15052126/d6dd00472e.pdf},
 issn = {2635-098X},
 journal = {Digital Discovery},
 month = {11},
 title = {Assay-Based Machine Learning: Rethinking Evaluation in Drug Discovery},
 url = {https://doi.org/10.1039/D6DD00472E},
 year = {2026}
}
DOI URL

Evaluation beyond goodness of fit: Quantifying biophysical alignment of AI models for kinase-centric drug discovery

2026 — Artificial Intelligence in the Life Sciences (AILSCI)

@article{grossEvaluationBeyondGoodnessOf2026,
 author = {Joschka Groß and Michael Backenköhler and Paula Linh Kramer and Raquel López-Ríos De Castro and Guillermo Pérez-Hernández and Verena Wolf and Andrea Volkamer},
 doi = {10.1016/j.ailsci.2026.100169},
 journal = {Artificial Intelligence in the Life Sciences (AILSCI)},
 month = {6},
 number = {article 100169},
 pages = {1-17},
 publisher = {Elsevier},
 title = {Evaluation beyond goodness of fit: Quantifying biophysical alignment of AI models for kinase-centric drug discovery},
 url = {https://www.sciencedirect.com/science/article/pii/S2667318526000176?via%3Dihub},
 volume = {9},
 year = {2026}
}
DOI URL

B-cos GNNs: Faithful Explanations through Dynamic Linearity

2026 — arXiv preprint arXiv:2605.19778

@article{grossBCosGNNsFaithfulExplanations2026,
 author = {Groß, Joschka and Solanki, Mohammad Shaique and Wolf, Verena},
 journal = {arXiv preprint arXiv:2605.19778},
 title = {B-cos GNNs: Faithful Explanations through Dynamic Linearity},
 year = {2026}
}

2025

1 paper

The Journey of Data: Lessons Learned in Modeling Kinase Affinity, Selectivity, and Resistance [Article v1. 0]

2025 — Living Journal of Computational Molecular Science (LiveCoMS)

PDF
@article{castroTheJourneyOfData2025,
 author = {Raquel López-Ríos de Castro and Jaime Rodríguez-Guerra and David Schaller and Talia K Kimber and Corey Taylor and Jessica B White and Michael Backenköhler and Joschka Groß and Alexander M Payne and Benjamin Kaminow and Ivan Pulido and Sukrit Singh and Paula Linh Kramer and Guilermo Pérez-Hernández and Andrea Volkamer and John D Chodera},
 doi = {10.33011/livecoms.6.1.3875},
 editor = {David L. Mobley and Michael R. Shirts and Daniel M. Zuckerman},
 journal = {Living Journal of Computational Molecular Science (LiveCoMS)},
 month = {11},
 number = {1},
 pages = {3875-3875},
 publisher = {University of Colorado Boulder, Boulder, Colorado, USA},
 title = {The Journey of Data: Lessons Learned in Modeling Kinase Affinity, Selectivity, and Resistance [Article v1. 0]},
 url = {https://www.dfki.de/fileadmin/user_upload/import/16917_The_Journey_of_Data_Lessons_Learned_in_Modeling_Ki.pdf},
 volume = {6},
 year = {2025}
}
DOI

2024

2 papers

Guided docking as a data generation approach facilitates structure-based machine learning on kinases

2024 — Journal of Chemical Information and Modeling (JCIM)

PDF
@article{backenkohlerGuidedDockingAsA2024,
 author = {Michael Backenköhler and Joschka Groß and Verena Wolf and Andrea Volkamer},
 doi = {10.1021/acs.jcim.4c00055},
 journal = {Journal of Chemical Information and Modeling (JCIM)},
 month = {5},
 number = {10},
 pages = {4009-4020},
 publisher = {ACS Publications, Washington DC},
 title = {Guided docking as a data generation approach facilitates structure-based machine learning on kinases},
 url = {https://pubs.acs.org/doi/10.1021/acs.jcim.4c00055},
 volume = {64},
 year = {2024}
}
DOI URL

Structural activity prediction models recover known binding modes (Poster abstract)

2024 — ICML'24 Workshop ML for Life and Material Science: From Theory to Industry Applications. International Conference on Machine Learning (ICML-2024)

@inproceedings{backenkohlerStructuralActivityPredictionModels2024,
 author = {Michael Backenköhler and Joschka Groß and Paula Linh Kramer and Verena Wolf and Andrea Volkamer},
 booktitle = {ICML'24 Workshop ML for Life and Material Science: From Theory to Industry Applications. International Conference on Machine Learning (ICML-2024)},
 publisher = {Openreview},
 title = {Structural activity prediction models recover known binding modes (Poster abstract)},
 url = {https://openreview.net/forum?id=9O27GTKm1B},
 year = {2024}
}
URL

2023

2 papers

DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning--Extended Version

2023 — ACM Transactions on Modeling and Computer Simulation (TOMACS)

@article{grosDSMCEvaluationStagesFostering2023,
 author = {Timo P. Gros and Joschka Groß and Daniel Höller and Jörg Hoffmann and Michaela Klauck and Hendrik Meerkamp and Nicola J. Müller and Lukas Schaller and Verena Wolf},
 doi = {10.1145/3607198},
 journal = {ACM Transactions on Modeling and Computer Simulation (TOMACS)},
 month = {10},
 number = {4},
 pages = {1-28},
 publisher = {ACM, New York, NY, USA},
 title = {DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning--Extended Version},
 volume = {33},
 year = {2023}
}
DOI

TeachOpenCADD goes Deep Learning: Open-source Teaching Platform Exploring Molecular DL Applications

2023 — ChemRxiv

PDF
@article{backenkohlerTeachOpenCADDGoesDeepLearning2023,
 author = {Michael Backenköhler and Paula Linh Kramer and Joschka Groß and Gerrit Großmann and Roman Joeres and Azat Tagirdzhanov and Dominique Sydow and Hamza Ibrahim and Floriane Odje and Verena Wolf and Andrea Volkamer},
 doi = {10.26434/chemrxiv-2023-kz1pb},
 journal = {ChemRxiv},
 month = {5},
 pages = {---},
 publisher = {ChemRxiv.org},
 title = {TeachOpenCADD goes Deep Learning: Open-source Teaching Platform Exploring Molecular DL Applications},
 url = {https://doi.org/10.26434/chemrxiv-2023-kz1pb},
 volume = {ChemRxiv Online},
 year = {2023}
}
DOI URL

2020

2 papers

Novel Is Not Always Better: On the Relation between Novelty and Dominance Pruning

2020 — Proceedings of the AAAI Conference on Artificial Intelligence

PDF
@article{grossNovelIsNotAlways2020,
 author = {Joschka Groß and Alvaro Torralba and Maximilian Fickert},
 doi = {10.1609/aaai.v34i06.6541},
 editor = {Vincent Conitzer and Fei Sha},
 journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
 month = {4},
 number = {06: AAAI-20 Technical Tracks 6},
 pages = {9875-9882},
 publisher = {AAAI Press, Palo Alto, California, USA},
 title = {Novel Is Not Always Better: On the Relation between Novelty and Dominance Pruning},
 url = {https://cdn.aaai.org/ojs/6541/6541-13-9766-1-10-20200519.pdf},
 volume = {34},
 year = {2020}
}
DOI

Real-Time Decision Making for a Car Manufacturing Process Using Deep Reinforcement Learning

2020 — 2020 Winter Simulation Conference (WSC). Winter Simulation Conference (WSC-2020), December 14-18, Orlando,, FL, USA

@inproceedings{grosRealTimeDecisionMakingFor2020,
 author = {Timo P. Gros and Joschka Groß and Verena Wolf},
 booktitle = {2020 Winter Simulation Conference (WSC). Winter Simulation Conference (WSC-2020), December 14-18, Orlando,, FL, USA},
 doi = {10.1109/WSC48552.2020.9383884},
 pages = {3032-3044},
 publisher = {IEEE},
 title = {Real-Time Decision Making for a Car Manufacturing Process Using Deep Reinforcement Learning},
 year = {2020}
}
DOI

Teaching (8)