Research Group Debarghya Ghoshdastidar
Debarghya Ghoshdastidar
Prof. Dr.
Core PI
Debarghya Ghoshdastidar
is Professor for Theoretical Foundations of Artificial Intelligence at TU Munich.
He conducts research in the theory of machine learning, artificial intelligence and network science. The main focus of his research is on the statistical understanding and interpretability of methods used in machine learning. His works provide new insights and algorithms for decision problems, involving complex data such as networks and preference relations, that arise in various fields including neuroscience, crowdsourcing and computer vision.
Team members @MCML
PhD Students
Maedeh Zarvandi
→ Group Debarghya Ghoshdastidar
Theoretical Foundations of Artificial Intelligence
Recent News @MCML
Publications @MCML
2026
[7]
M. Zarvandi • M. Timothy • T. Wasserer • D. Ghoshdastidar
Interpretable Self-Supervised Learning via Representer Landmarks and Nyström Approximation.
ICML 2026 - 43rd International Conference on Machine Learning. Seoul, South Korea, Jul 06-11, 2026. To be published. Preprint available. URL
Interpretable Self-Supervised Learning via Representer Landmarks and Nyström Approximation.
ICML 2026 - 43rd International Conference on Machine Learning. Seoul, South Korea, Jul 06-11, 2026. To be published. Preprint available. URL
[6]
Y. Han • Y. Wang • D. Ghoshdastidar • J. Kinder
ALPHA: Active Learning with PAC-Bayesian Theory for Android Malware Detection.
ACM ASIACSS 2026 - 21st ACM ASIA Conference on Computer and Communications Security. Bangalore, India, Jun 01-05, 2026. DOI GitHub
ALPHA: Active Learning with PAC-Bayesian Theory for Android Malware Detection.
ACM ASIACSS 2026 - 21st ACM ASIA Conference on Computer and Communications Security. Bangalore, India, Jun 01-05, 2026. DOI GitHub
[5]
A. Mohgaonkar • L. Gosch • M. Sabanayagam • D. Ghoshdastidar • S. Günnemann
Exact Certification of Neural Networks and Partition Aggregation Ensembles against Label Poisoning.
Trustworthy AI @ICLR 2026 - Workshop on Principled Design for Trustworthy AI - Interpretability, Robustness, and Safety across Modalities at the 14th International Conference on Learning Representations. Rio de Janeiro, Brazil, Apr 23-27, 2026. To be published. Preprint available. arXiv GitHub
Exact Certification of Neural Networks and Partition Aggregation Ensembles against Label Poisoning.
Trustworthy AI @ICLR 2026 - Workshop on Principled Design for Trustworthy AI - Interpretability, Robustness, and Safety across Modalities at the 14th International Conference on Learning Representations. Rio de Janeiro, Brazil, Apr 23-27, 2026. To be published. Preprint available. arXiv GitHub
2025
[4]
A. Crăciun • D. Ghoshdastidar
Non-Singularity of the Gradient Descent Map for Neural Networks with Piecewise Analytic Activations.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. URL
Non-Singularity of the Gradient Descent Map for Neural Networks with Piecewise Analytic Activations.
NeurIPS 2025 - 39th Conference on Neural Information Processing Systems. San Diego, CA, USA, Nov 30-Dec 07, 2025. Spotlight Presentation. URL
[3]
P. Esser • M. Fleissner • D. Ghoshdastidar
Theoretical Foundations of Representation Learning using Unlabeled Data: Statistics and Optimization.
Preprint (Sep. 2025). arXiv
Theoretical Foundations of Representation Learning using Unlabeled Data: Statistics and Optimization.
Preprint (Sep. 2025). arXiv
[2]
L. Gosch • M. Sabanayagam • D. Ghoshdastidar • S. Günnemann
Provable Robustness of (Graph) Neural Networks Against Data Poisoning and Backdoor Attacks.
Transactions on Machine Learning Research Vol. 2025. Jun. 2025. URL
Provable Robustness of (Graph) Neural Networks Against Data Poisoning and Backdoor Attacks.
Transactions on Machine Learning Research Vol. 2025. Jun. 2025. URL
[1]
M. Sabanayagam • L. Gosch • S. Günnemann • D. Ghoshdastidar
Exact Certification of (Graph) Neural Networks Against Label Poisoning.
VerifAI @ICLR 2025 - Workshop on AI Verification in the Wild at the 13th International Conference on Learning Representations. Singapore, Apr 24-28, 2025. URL GitHub
Exact Certification of (Graph) Neural Networks Against Label Poisoning.
VerifAI @ICLR 2025 - Workshop on AI Verification in the Wild at the 13th International Conference on Learning Representations. Singapore, Apr 24-28, 2025. URL GitHub
©all images: LMU | TUM
Back to Top
2025-03-17 - Last modified: 2026-07-03