Publications
- 2024 -
[arXiv] Hyperparameter Optimization in Machine Learning. Luca Franceschi, Michele Donini, Valerio Perrone, Aaron Klein, Cédric Archambeau, Matthias Seeger, Massimiliano Pontil, Paolo Frasconi.
[C34] Explaining Probabilistic Models with Distributional Values. Luca Franceschi, Michele Donini, Cédric Archambeau, Matthias Seeger. ICML 2024. Accepted as Spotlight Paper (acceptance rate 3.5%).
[arXiv] Evaluating Large Language Models with fmeval. Pola Schwöbel, Luca Franceschi, Muhammad Bilal Zafar, Keerthan Vasist, Aman Malhotra, Tomer Shenhar, Pinal Tailor, Pinar Yilmaz, Michael Diamond, Michele Donini.
- 2023 -
[C33] Geographical Erasure in Language Generation. Pola Schwöbel, Jacek Golebiowski, Michele Donini, Cédric Archambeau, Danish Pruthi. EMNLP 2023.
[W12] Explaining Multiclass Classifiers with Categorical Values: A Case Study in Radiography. Luca Franceschi, Bilal Zafar, Matthias Seeger, Gianluca Detommaso, Cemre Zor, Cédric Archambeau, Michele Donini. TMLH Workshop - ICLR 2023.
[C32] Efficient fair PCA for fair representation learning. Matthäus Kleindessner, Michele Donini, Chris Russell, Bilal Zafar. AISTAT 2023.
[J13] Fortuna: A Library for Uncertainty Quantification in Deep Learning. Gianluca Detommaso, Alberto Gasparin, Michele Donini, Matthias Seeger, Andrew Gordon Wilson, Cédric Archambeau. JMLR.
- 2022 -
[C31] Amazon SageMaker Model Monitor: A System for Real-Time Insights into Deployed Machine Learning Models. David Nigenda, Zohar Karnin, Muhammad Bilal Zafar, Raghu Ramesha, Alan Tan, Michele Donini, Krishnaram Kenthapadi. KDD 2022.
[arXiv] Diverse Counterfactual Explanations for Anomaly Detection in Time Series. Deborah Sulem, Michele Donini, Muhammad Bilal Zafar, Francois-Xavier Aubet, Jan Gasthaus, Tim Januschowski, Sanjiv Das, Krishnaram Kenthapadi, Cédric Archambeau.
[J12] FinLex: An Effective Use of Word Embeddings for Financial Lexicon Generation. Sanjiv Das, Michele Donini, Muhammad Bilal Zafar, John He, Krishnaram Kenthapadi. The Journal of Finance and Data Science. Outstanding Article Award 2022.
- 2021 -
[arXiv] More Than Words: Towards Better Quality Interpretations of Text Classifiers. Muhammad Bilal Zafar, Philipp Schmidt, Michele Donini, Cédric Archambeau, Felix Biessmann, Sanjiv Das, Krishnaram Kenthapadi.
[J11] Fairness measures for machine learning in finance. Sanjiv Das, Michele Donini, Jason Gelman, Kevin Haas, Mila Hardt, Jared Katzman, Krishnaram Kenthapadi, Pedro Larroy, Pinar Yilmaz, Bilal Zafar. The Journal of Financial Data Science.
[arXiv] Multi-objective Asynchronous Successive Halving. Robin Schmucker, Michele Donini, Muhammad Bilal Zafar, David Salinas, Cédric Archambeau.
[J10] Deep Fair Models for Complex Data: Graphs Labeling and Explainable Face Recognition. Danilo Franco, Nicolò Navarin, Michele Donini, Davide Anguita, Luca Oneto. Neurocomputing.
[C30] Amazon SageMaker Automatic Model Tuning: Scalable Gradient-Free Optimization. Valerio Perrone, Huibin Shen, Aida Zolic, Iaroslav Shcherbatyi, Amr Ahmed, Tanya Bansal, Michele Donini, Fela Winkelmolen, Rodolphe Jenatton, Jean Baptiste Faddoul, Barbara Pogorzelska, Miroslav Miladinovic, Krishnaram Kenthapadi, Matthias Seeger, Cédric Archambeau. KDD 2021.
[C29] Amazon SageMaker Clarify: Machine Learning Bias Detection and Explainability in the Cloud. Michaela Hardt, Xiaoguang Chen, Xiaoyi Cheng, Michele Donini, Jason Gelman, Satish Gollaprolu, John He, Pedro Larroy, Xinyu Liu, Nick McCarthy, Ashish Rathi, Scott Rees, Ankit Siva, ErhYuan Tsai, Keerthan Vasist, Pinar Yilmaz, Muhammad Bilal Zafar, Sanjiv Das, Kevin Haas, Tyler Hill, Krishnaram Kenthapadi. KDD 2021.
[C28] On the Lack of Robust Interpretability of Transformer-based Text Classifiers. Muhammad Bilal Zafar, Michele Donini, Dylan Slack, Cédric Archambeau, Sanjiv Das, Krishnaram Kenthapadi. ACL-IJCNLP 2021.
[J9] Voting with Random Classifiers (VORACE): Theoretical and Experimental Analysis. Cristina Cornelio, Michele Donini, Andrea Loreggia, Maria Silvia Pini, Francesca Rossi. Autonomous Agents and Multi-Agent Systems, Springer Journals.
[C27] Fair Bayesian Optimization. Valerio Perrone, Michele Donini, Muhammad Bilal Zafar, Robin Schmucker, Krishnaram Kenthapadi, Cédric Archambeau. AIES 2021.
- 2020 -
[W11] Multi-Objective Multi-Fidelity Hyperparameter Optimization with Application to Fairness. Robin Schmucker, Michele Donini, Valerio Perrone, Muhammad Bilal Zafar, Cédric Archambeau. MetaLearn, NeurIPS 2020 workshop.
[C26] Exploiting MMD and Sinkhorn Divergences for Learning Fair and Transferable Representations. Luca Oneto, Michele Donini, Giulia Luise, Carlo Ciliberto, Andreas Maurer and Massimiliano Pontil. NeurIPS 2020.
[C25] Learning Fair and Transferable Representations with Theoretical Guarantees. Luca Oneto, Michele Donini, Massimiliano Pontil and Andreas Maurer. DSAA 2020.
[W10] Bayesian Optimization with Fairness Constraints. Valerio Perrone, Michele Donini, Krishnaram Kenthapadi and Cédric Archambeau. AutoML, ICML 2020 workshop. Best paper award.
[C24] MARTHE: Scheduling the Learning Rate Via Online Hypergradients. Michele Donini, Luca Franceschi, Massimiliano Pontil, Orchid Majumder, Paolo Frasconi. IJCAI2020.
[C23] General Fair Empirical Risk Minimization. Luca Oneto, Michele Donini, Massimiliano Pontil. IJCNN 2020.
[C22] Learning Deep Fair Graph Neural Networks. Nicolò Navarin, Luca Oneto and Michele Donini. ESANN 2020.
[C21] Voting with Random Classifiers (VORACE). Cristina Cornelio, Michele Donini, Andrea Loreggia, Maria Silvia Pini and Francesca Rossi. AAMAS 2020.
[J8] Randomized Learning and Generalization of Fair and Private Classifiers: from PAC-Bayes to Stability and Differential Privacy. Luca Oneto, Michele Donini, Massimiliano Pontil and John Shawe-Taylor. Neurocomputing.
- 2019 -
[W9] Learning Fair and Transferable Representations. Luca Oneto, Michele Donini, Massimiliano Pontil and Andreas Maurer. HCML, NeurIPS 2019 workshop.
[W8] Learning the Learning Rate for Gradient Descent by Gradient Descent. Orchid Majumder, Michele Donini and Pratik Chaudhari. AutoML, ICML 2019 workshop.
[J7] Combining heterogeneous data sources for prediction: re-weighting and selecting what is important. Michele Donini, João M. Monteiro, Massimiliano Pontil, Tim Hahn, Andreas J. Fallgatter, John Shawe-Taylor and Janaina Mourao-Miranda. NeuroImage.
[C20] PAC-Bayes and Fairness: Risk and Fairness Bounds on Distribution Dependent Fair Priors. Luca Oneto, Michele Donini and Massimiliano Pontil. ESANN 2019.
[C19] Taking Advantage of Multitask Learning for Fair Classification. Luca Oneto, Michele Donini, Amon Elder and Massimiliano Pontil. AIES 2019.
- 2018 -
[W7] Voting with Random Neural Networks: a Democratic Ensemble Classifier. Michele Donini, Andrea Loreggia, Maria Silvia Pini and Francesca Rossi. RiCeRcA Workshop - 17th International Conference of the Italian Association for Artificial Intelligence 2018.
[C18] Empirical Risk Minimization Under Fairness Constraints. Michele Donini, Luca Oneto, Shai Ben-David, John Shawe-Taylor and Massimiliano Pontil. NeurIPS 2018.
[C17] Emerging trends in machine learning: beyond conventional methods and data. Luca Oneto, Nicolò Navarin, Michele Donini and Davide Anguita. ESANN 2018.
[J6] Scuba: scalable kernel-based gene prioritization. Guido Zampieri, Dinh Van Tran, Michele Donini, Nicolò Navarin, Fabio Aiolli, Alessandro Sperduti and Giorgio Valle. BMC Bioinformatics.
- 2017 -
[W6] A bridge between hyperparameter optimization and learning-to-learn. Luca Franceschi, Paolo Frasconi, Michele Donini and Massimiliano Pontil. MetaML2017, NIPS 2017 workshop.
[W5] An Efficient Method to Impose Fairness in Linear Models. Michele Donini, Shai Ben-David, Massimiliano Pontil and John Shawe-Taylor. WPOC2017, NIPS 2017 workshop.
[C16] A Speaker Adaptive DNN Training Approach for Speaker-independent Acoustic Inversion. Leonardo Badino, Luca Franceschi, Michele Donini and Massimiliano Pontil. Interspeech 2017.
[C15] Forward and Reverse Gradient-Based Hyperparameter Optimization. Luca Franceschi, Michele Donini, Paolo Frasconi and Massimiliano Pontil. ICML 2017.
[W4] On Hyperparameter Optimization in Learning Systems. Luca Franceschi, Michele Donini, Paolo Frasconi and Massimiliano Pontil. ICLR 2017.
[C14] Fast Hyperparameter Selection for Graph Kernels via Subsampling and Multiple Kernel Learning. Michele Donini, Nicolò Navarin, Ivano Lauriola, Fabio Aiolli and Fabrizio Costa. ESANN 2017.
[C13] Learning Dot Product Polynomials for multiclass problems. Ivano Lauriola, Michele Donini and Aiolli Fabio. ESANN 2017.
[W3] Berhu Penalty for Matrix and Tensor Estimation. Giulia Denevi, Michele Donini, Massimiliano Pontil. Signal Processing with Adaptive Sparse Structured Representations (SPARS) workshop, 2017.
[J5] Learning with Kernels: a Local Rademacher Complexity-based Analysis with Application to Graph Kernels. Luca Oneto, Nicolò Navarin, Michele Donini, Alessandro Sperduti, Fabio Aiolli and Davide Anguita. IEEE Transactions on Neural Networks and Learning Systems.
[J4] Measuring the Expressivity of Graph Kernels through Statistical Learning Theory. Luca Oneto, Nicolò Navarin, Michele Donini, Alessandro Sperduti, Fabio Aiolli and Davide Anguita. Neurocomputing.
- 2016 -
[C12] A Multimodal Multiple Kernel Learning Approach to Alzheimer’s Disease Detection. Michele Donini, João M. Monteiro, Massimiliano Pontil, John Shawe-Taylor and Janaina Mourao-Miranda. MLSP 2016.
[C11] Kaczmarz and Cimmino: iterative and layer-oriented approaches to atmospheric tomography. Chiara Garbellotto, Michele Donini, Roberto Ragazzoni, Carmelo Arcidiacono, Andrea Baruffolo and Jacopo Farinato. SPIE 2016.
[C10] Distributed Variance Regularized Multitask Learning. Michele Donini, David Martinez-Rego, Martin Goodson, John Shawe-Taylor and Massimiliano Pontil. IJCNN 2016.
[J3] Learning Deep Kernels in the Space of Dot Product Polynomials. Michele Donini and Fabio Aiolli. Machine Learning Journal.
[J2] Stairstep recognition and counting in a serious Game for increasing users’ physical activity. Matteo Ciman, Michele Donini, Ombretta Gaggi and Fabio Aiolli. Personal and Ubiquitous Computing.
[C9] Advances in Learning with Kernels: Theory and Practice in a World of growing Constraints. Luca Oneto, Nicolò Navarin, Michele Donini, Fabio Aiolli and Davide Anguita. ESANN 2016.
[C8] Measuring the Expressivity of Graph Kernels through the Rademacher Complexity. Oneto Luca, Navarin Nicolò, Donini Michele, Sperduti Alessandro, Aiolli Fabio and Anguita Davide. ESANN 2016.
- 2015 -
[C7] Multiple Graph-Kernel Learning. Fabio Aiolli, Michele Donini, Nicoló Navarin and Alessandro Sperduti. SSCI 2015.
[C6] Feature and kernel learning. Verónica Bolón-Canedo, Michele Donini and Fabio Aiolli. ESANN 2015.
[J1] EasyMKL: a scalable multiple kernel learning algorithm. Fabio Aiolli and Michele Donini. Neurocomputing.
- 2014 -
[C5] ClimbTheWorld: Real-time stairstep counting to increasephysical activity. Fabio Aiolli, Matteo Ciman, Michele Donini and Ombretta Gaggi. MOBIQUITOUS 2014.
[C4] Learning Anisotropic RBF Kernels. Fabio Aiolli and Michele Donini. ICANN 2014.
[C3] Easy multiple kernel learning. Fabio Aiolli and Michele Donini. ESANN 2014.
[C2] A Serious Game to persuade people to use stairs. Fabio Aiolli, Matteo Ciman, Michele Donini and Ombretta Gaggi. Persuasive 2014.
- 2013 -
[W2] Voting for Classifier Selection. Michele Donini and Maria Silvia Pini. MLDM.it 2013.
[C1] Stacking Models for Efficient Annotation of Brain Tissues in MR Volumes. Fabio Aiolli, Michele Donini, Enea Poletti and Enrico Grisan. Medicon 2013.
[W1] Designing quiet rotorcraft landing trajectories with probabilistic road maps. Robert A. Morris, Michele Donini, K. Brent Venable and Matthew Johnson. SPARK 2013.