Publications

2026

1. From Privacy to Generalization: Linear Max-Information Bounds for DP-SGD

   Christoph H. Lampert ,  and  Hossein Zakerinia

   arXiv preprint arXiv:2605.26222, 2026

   Bib PDF

   @article{Lampert2026privacy,
     title = {From Privacy to Generalization: Linear Max-Information Bounds for DP-SGD},
     author = {Lampert, Christoph H. and Zakerinia, Hossein},
     journal = {arXiv preprint arXiv:2605.26222},
     year = {2026},
   }

2. Federated Learning with Unlabeled Clients: Personalization Can Happen in Low Dimensions

   Hossein Zakerinia ,  Jonathan Scott ,  and  Christoph H. Lampert

   In International Conference on Machine Learning (ICML) , 2026

   Bib PDF

   @inproceedings{zakerinia2025FLowDUP,
     title = {Federated Learning with Unlabeled Clients: Personalization Can Happen in Low Dimensions},
     author = {Zakerinia, Hossein and Scott, Jonathan and Lampert, Christoph H.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2026},
   }

2025

1. Fast Rate Bounds for Multi-Task and Meta-Learning with Different Sample Sizes

   Hossein Zakerinia ,  and  Christoph H. Lampert

   In Conference on Neural Information Processing Systems (NeurIPS) , 2025

   Bib PDF

   @inproceedings{zakerinia2025fast,
     title = {Fast Rate Bounds for Multi-Task and Meta-Learning with Different Sample Sizes},
     author = {Zakerinia, Hossein and Lampert, Christoph H.},
     booktitle = {Conference on Neural Information Processing Systems (NeurIPS)},
     year = {2025},
   }

2. From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning

   Hossein Zakerinia ,  Dorsa Ghobadi ,  and  Christoph H. Lampert

   arXiv preprint arXiv:2501.19067, 2025

   Bib PDF

   @article{zakerinia2025deepMTL,
     title = {From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning},
     author = {Zakerinia, Hossein and Ghobadi, Dorsa and Lampert, Christoph H.},
     journal = {arXiv preprint arXiv:2501.19067},
     year = {2025},
   }

3. DP-KAN: Differentially Private Kolmogorov-Arnold Networks

   arXiv preprint arXiv:2407.12569, 2025

   Bib PDF

   @article{Kalinin2025DPKAN,
     title = {DP-KAN: Differentially Private Kolmogorov-Arnold Networks},
     author = {},
     journal = {arXiv preprint arXiv:2407.12569},
     year = {2025},
   }

2024

1. More Flexible PAC-Bayesian Meta-Learning by Learning Learning Algorithms

   Hossein Zakerinia ,  Amin Behjati ,  and  Christoph H. Lampert

   In International Conference on Machine Learning (ICML) , 2024

   Bib PDF

   @inproceedings{zakerinia2024flexible,
     title = {More Flexible PAC-Bayesian Meta-Learning by Learning Learning Algorithms},
     author = {Zakerinia, Hossein and Behjati, Amin and Lampert, Christoph H.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2024},
   }

2. Communication-Efficient Federated Learning With Data and Client Heterogeneity

   Hossein Zakerinia ,  Shayan Talaei ,  Giorgi Nadiradze , and 1 more author

   In International Conference on Artificial Intelligence and Statistics (AISTATS) , 2024

   Bib PDF

   @inproceedings{zakerinia2024quafl,
     title = {Communication-Efficient Federated Learning With Data and Client Heterogeneity},
     author = {Zakerinia, Hossein and Talaei, Shayan and Nadiradze, Giorgi and Alistarh, Dan},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2024},
   }

3. PeFLL: Personalized Federated Learning by Learning to Learn

   Jonathan Scott ,  Hossein Zakerinia ,  and  Christoph H. Lampert

   In International Conference on Learning Representations (ICLR) , 2024

   Bib PDF

   @inproceedings{scott2024pefll,
     title = {PeFLL: Personalized Federated Learning by Learning to Learn},
     author = {Scott, Jonathan and Zakerinia, Hossein and Lampert, Christoph H.},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2024},
   }

2021

1. Mg-bert: Multi-graph augmented bert for masked language modeling

   Parishad BehnamGhader ,  Hossein Zakerinia ,  and  Mahdieh Soleymani Baghshah

   In Proceedings of the Fifteenth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-15) , 2021

   Bib PDF

   @inproceedings{behnamghader2021mg,
     title = {Mg-bert: Multi-graph augmented bert for masked language modeling},
     author = {BehnamGhader, Parishad and Zakerinia, Hossein and Baghshah, Mahdieh Soleymani},
     booktitle = {Proceedings of the Fifteenth Workshop on Graph-Based Methods for Natural Language Processing (TextGraphs-15)},
     year = {2021},
   }