Publications

2026

1. 

   Hot Training Errors Predict Cold Test Errors in the Interpolation Regime

   Erfan Mirzaei, Andreas Maurer, and Massimiliano Pontil

   In AXIOM: Foundations of Efficient Deep Learning Workshop at NeurIPS, 2026

   Bib HTML PDF

   @inproceedings{mirzaei2026hot,
     title = {Hot Training Errors Predict Cold Test Errors in the Interpolation Regime},
     author = {Mirzaei, Erfan and Maurer, Andreas and Pontil, Massimiliano},
     booktitle = {AXIOM: Foundations of Efficient Deep Learning Workshop at NeurIPS},
     year = {2026},
   }

2. 

   Generalization of Gibbs and Langevin Monte Carlo Algorithms in the Interpolation Regime

   Andreas Maurer, Erfan Mirzaei, and Massimiliano Pontil

   In Proceedings of the 43rd International Conference on Machine Learning, 2026

   arXiv Bib PDF Code

   @inproceedings{maurer2026generalization,
     title = {Generalization of {G}ibbs and {L}angevin {M}onte {C}arlo Algorithms in the Interpolation Regime},
     author = {Maurer, Andreas and Mirzaei, Erfan and Pontil, Massimiliano},
     booktitle = {Proceedings of the 43rd International Conference on Machine Learning},
     pages = {87068--87096},
     year = {2026},
     volume = {306},
     series = {Proceedings of Machine Learning Research},
     publisher = {PMLR},
   }

2025

1. 

   An Empirical Bernstein Inequality for Dependent Data in Hilbert Spaces and Applications

   Erfan Mirzaei, Andreas Maurer, Vladimir Kostic, and Massimiliano Pontil

   In International Conference on Artificial Intelligence and Statistics (AISTATS), 2025

   arXiv Bib PDF Code

   @inproceedings{mirzaei2025empirical,
     title = {An Empirical {B}ernstein Inequality for Dependent Data in {H}ilbert Spaces and Applications},
     author = {Mirzaei, Erfan and Maurer, Andreas and Kostic, Vladimir and Pontil, Massimiliano},
     booktitle = {International Conference on Artificial Intelligence and Statistics (AISTATS)},
     year = {2025},
   }

2. 

   kooplearn: A Scikit-Learn Compatible Library of Algorithms for Evolution Operator Learning

   Giacomo Turri, Grégoire Pacreau, Giacomo Meanti, Timothée Devergne, and 7 more authors

   arXiv preprint arXiv:2512.21409, 2025

   arXiv Bib PDF Code

   @article{turri2025kooplearn,
     title = {kooplearn: A Scikit-Learn Compatible Library of Algorithms for Evolution Operator Learning},
     author = {Turri, Giacomo and Pacreau, Gr{\'e}goire and Meanti, Giacomo and Devergne, Timoth{\'e}e and Ordonez, Daniel and Mirzaei, Erfan and Belucci, Bruno and Lounici, Karim and Kostic, Vladimir and Pontil, Massimiliano and others},
     journal = {arXiv preprint arXiv:2512.21409},
     year = {2025},
   }

3. 

   Exploiting Expertise of Non-Expert and Diverse Agents in Social Bandit Learning: A Free Energy Approach

   Erfan Mirzaei, Seyed Pooya Shariatpanahi, Alireza Tavakoli, Reshad Hosseini, and 1 more author

   IEEE Transactions on Cognitive and Developmental Systems, 2025

   arXiv Bib PDF

   @article{mirzaei2025exploiting,
     title = {Exploiting Expertise of Non-Expert and Diverse Agents in Social Bandit Learning: A Free Energy Approach},
     author = {Mirzaei, Erfan and Shariatpanahi, Seyed Pooya and Tavakoli, Alireza and Hosseini, Reshad and Ahmadabadi, Majid Nili},
     journal = {IEEE Transactions on Cognitive and Developmental Systems},
     year = {2025},
     publisher = {IEEE},
   }

2022

1. 

   BioLCNet: Reward-modulated locally connected spiking neural networks

   Hafez Ghaemi, Erfan Mirzaei, Mahbod Nouri, and Saeed Reza Kheradpisheh

   In International Conference on Machine Learning, Optimization, and Data Science, 2022

   arXiv Bib PDF Code

   @inproceedings{ghaemi2022biolcnet,
     title = {BioLCNet: Reward-modulated locally connected spiking neural networks},
     author = {Ghaemi, Hafez and Mirzaei, Erfan and Nouri, Mahbod and Kheradpisheh, Saeed Reza},
     booktitle = {International Conference on Machine Learning, Optimization, and Data Science},
     pages = {564--578},
     year = {2022},
     organization = {Springer},
   }