About Me

I'm an assistant professor (RTD-A) at the Department of Electronics, Information and Bioengineering (DEIB) at Politecnico di Milano. I received my PhD in computer science at Politecnico di Milano, where I was advised by Prof. Nicola Gatti. My current research focuses on Artificial Intelligence and algorithmic game theory. In particular, I'm interested in combining machine learning techniques with economic paradigms to build strategic agents able to act in complex multi-agent environments.

More details in my CV.

Contacts:
Email: matteo DOT castiglioni AT polimi.it

Publications

Conference Papers
  • Hiring for An Uncertain Task: Joint Design of Information and Contracts
    Castiglioni Matteo, Chen Junjie
    ACM-SIAM Symposium on Discrete Algorithms, SODA 2025, New Orleans, USA
    [arXiv]
  • A Reduction from Multi-Parameter to Single-Parameter Bayesian Contract Design
    Castiglioni Matteo, Chen Junjie, Li Minming, Xu Haifeng, Zuo Song
    ACM-SIAM Symposium on Discrete Algorithms, SODA 2025, New Orleans, USA
    [arXiv]
  • Beyond Primal-Dual Methods in Bandits with Stochastic and Adversarial Constraints
    Bernasconi Martino, Castiglioni Matteo, Celli Andrea, Fusco Federico
    38th Conference on Neural Information Processing Systems, NeurIPS 2024, Vancouver, Canada
    [arXiv]
  • Online Bayesian Persuasion Without a Clue
    Bacchiocchi Francesco, Bollini Matteo, Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    38th Conference on Neural Information Processing Systems, NeurIPS 2024, Vancouver, Canada
  • Bandits with Ranking Feedback
    Maran Davide, Bacchiocchi Francesco, Stradi Francesco, Castiglioni Matteo, Gatti Nicola, Restelli Marcello
    38th Conference on Neural Information Processing Systems, NeurIPS 2024, Vancouver, Canada
  • Agent-Designed Contracts: How to Sell Hidden Actions
    Bernasconi Martino, Castiglioni Matteo, Celli Andrea
    The 25th ACM Conference on Economics and Computation, EC 2024, New Haven, USA
    [arXiv]
  • Multi-Agent Contract Design beyond Binary Actions
    Cacciamani Federico, Bernasconi Martino, Castiglioni Matteo, Gatti Nicola
    The 25th ACM Conference on Economics and Computation, EC 2024, New Haven, USA
    [arXiv]
  • Online Learning under Budget and ROI Constraints via Weak Adaptivity
    Castiglioni Matteo, Celli Andrea, Kroer Christian
    41st International Conference on Machine Learning, ICML 2024, Vienna, Austria
    [arXiv]
  • Online Learning in CMDPs: Handling Stochastic and Adversarial Constraints
    Stradi Francesco, Germano Jacopo, Genalti Gianmarco, Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    41st International Conference on Machine Learning, ICML 2024, Vienna, Austria
    [arXiv]
  • Graph-Triggered Rising Bandits
    Genalti Gianmarco, Mussi Marco, Gatti Nicola, Restelli Marcello, Castiglioni Matteo, Metelli Alberto
    41st International Conference on Machine Learning, ICML 2024, Vienna, Austria
    [paper]
  • No-Regret Learning in Bilateral Trade via Global Budget Balance
    Bernasconi Martino, Castiglioni Matteo, Celli Andrea, Fusco Federico
    56th Annual ACM Symposium on Theory of Computing, STOC 2024, Vancouver, Canada
    [arXiv]
  • Learning Optimal Contracts: How to Exploit Small Action Spaces
    Bacchiocchi Francesco, Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    12th International Conference on Learning Representations, ICLR 2024, Vienna, Austria
    [arXiv]
  • Online Information Acquisition: Hiring Multiple Agents
    Cacciamani Federico, Castiglioni Matteo, Gatti Nicola
    12th International Conference on Learning Representations, ICLR 2024, Vienna, Austria
    [arXiv]
  • Bandits with Replenishable Knapsacks: the Best of both Worlds
    Bernasconi Martino, Castiglioni Matteo, Celli Andrea, Fusco Federico
    12th International Conference on Learning Representations, ICLR 2024, Vienna, Austria
    [arXiv]
  • Finding Effective Ad Allocations: How to Exploit User History
    Castiglioni Matteo, Latino Alberto, Marchesi Alberto, Romano Giulia, Gatti Nicola, Palayamkottai Chokha
    23rd International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2024, Auckland, New Zealand
  • Maximizing Revenue from Selfish Agents in Crowd Tasks: Indirect Incentive Strategies
    Montazeri Mina, Castiglioni Matteo, Romano Giulia, Hamed Kebriaei, Gatti Nicola
    63rd IEEE Conference on Decision and Control , CDC 2024, Milan, Italy
  • Persuading Farsighted Receivers in MDPs: the Power of Honesty
    Bernasconi Martino, Castiglioni Matteo, Marchesi Alberto, Mutti Mirco
    36th Conference on Neural Information Processing Systems, NeurIPS 2023, New Orleans, USA
    [paper]
  • Multi-Agent Contract Design: How to Commission Multiple Agents with Individual Outcome
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    The 24th ACM Conference on Economics and Computation, EC 2023, London, UK
    [arXiv]
  • Online Mechanism Design for Information Acquisition
    Cacciamani Federico, Castiglioni Matteo, Gatti Nicola
    The 40th International Conference on Machine Learning, ICML 2023, Honolulu, USA
    [arXiv]
  • Constrained Phi-Equilibria
    Bernasconi Martino, Castiglioni Matteo, Marchesi Alberto, Trovò Francesco, Gatti Nicola
    The 40th International Conference on Machine Learning, ICML 2023, Honolulu, USA
    [arXiv]
  • Optimal Rates and Efficient Algorithms for Online Bayesian Persuasion
    Bernasconi Martino, Castiglioni Matteo, Celli Andrea, Marchesi Alberto, Gatti Nicola, Trovò Francesco
    The 40th International Conference on Machine Learning, ICML 2023, Honolulu, USA
    [arXiv]
  • A Unifying Framework for Online Optimization with Long-Term Constraints
    Castiglioni Matteo, Celli Andrea, Marchesi Alberto, Romano Giulia, Gatti Nicola
    36th Conference on Neural Information Processing Systems, NeurIPS 2022, New Orleans, USA
    [arXiv]
  • Sequential Information Design: Learning to Persuade in the Dark
    Bernasconi Martino, Castiglioni Matteo, Marchesi Alberto, Gatti Nicola, Trovò Francesco
    36th Conference on Neural Information Processing Systems, NeurIPS 2022, New Orleans, USA
    [arXiv]
  • Safe Learning in Tree-Form Sequential Decision Making: Handling Hard and Soft Constraints
    Bernasconi Martino, Cacciamani Federico, Castiglioni Matteo, Marchesi Alberto, Gatti Nicola, Trovò Francesco
    The 39th International Conference on Machine Learning, ICML 2022, Baltimora, USA
    [paper]
  • Online Learning with Knapsacks: the Best of Both Worlds
    Castiglioni Matteo, Celli Andrea, Kroer Christian
    The 39th International Conference on Machine Learning, ICML 2022, Baltimora, USA
    [paper] [arXiv]
  • Designing Menus of Contracts Efficiently: the Power of Randomization
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    The 23rd ACM Conference on Economics and Computation, EC 2022, Boulder, USA
    [paper] [arXiv]
  • The Power of Media Agencies in Ad Auctions: Improving Utility through Coordinated Bidding
    Romano Giulia, Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    31st International Joint Conference on Artificial Intelligence, IJCAI 2022, Vienna, Austria
    [paper] [arXiv]
  • Public Signaling in Bayesian Ad Auctions
    Bacchiocchi Francesco, Castiglioni Matteo, Marchesi Alberto, Romano Giulia, Gatti Nicola
    31st International Joint Conference on Artificial Intelligence, IJCAI 2022, Vienna, Austria
    [paper] [arXiv]
  • Bayesian Persuasion Meets Mechanism Design: Going Beyond Intractability with Type Reporting
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    The 21st International Conference on Autonomous Agents and Multiagent Systems, AAMAS 2022, Virtual conference, Worldwide
    [paper] [arXiv]
  • Signaling in Posted Price Auctions
    Castiglioni Matteo, Romano Giulia, Marchesi Alberto, Gatti Nicola
    The 36th AAAI Conference on Artificial Intelligence, AAAI 2022, Virtual conference, Worldwide
    [paper] [arXiv]
  • Efficiency of Ad Auctions with Price Displaying
    Castiglioni Matteo, Ferraioli Diodato, Gatti Nicola, Marchesi Alberto, Romano Giulia
    The 36th AAAI Conference on Artificial Intelligence, AAAI 2022, Virtual conference, Worldwide
    [paper] [arXiv]
  • Multi-Receiver Online Bayesian Persuasion
    Castiglioni Matteo, Marchesi Alberto, Celli Andrea, Gatti Nicola
    The 38th International Conference on Machine Learning, ICML 2021, Virtual conference, Worldwide
    [paper]
  • Bayesian Agency: Linear Versus Tractable Contracts
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    The 22nd ACM Conference on Economics and Computation, EC 2021, Virtual conference, Worldwide
    [paper][arXiv]
  • Signaling in Bayesian Network Congestion Games: the Subtle Power of Symmetry
    Castiglioni Matteo, Celli Andrea, Marchesi Alberto, Gatti Nicola
    The 35th AAAI Conference on Artificial Intelligence, AAAI 2021, Virtual conference, Worldwide
    [paper]
  • Persuading Voters in District-based Elections
    Castiglioni Matteo, Gatti Nicola
    The 35th AAAI Conference on Artificial Intelligence, AAAI 2021, Virtual conference, Worldwide
    [paper]
  • Online Bayesian Persuasion
    Castiglioni Matteo, Celli Andrea, Marchesi Alberto, Gatti Nicola
    34th Conference on Neural Information Processing Systems, NeurIPS 2020, Virtual conference, Worldwide.
    [paper]
  • Election Control in Social Networks via Edge Addition or Removal
    Castiglioni Matteo, Ferraioli Diodato, Gatti Nicola
    34th AAAI Conference on Artificial Intelligence, AAAI 2020, New York, USA
    [paper]
  • Persuading Voters: It’s Easy to Whisper, It’s Hard to Speak Loud
    Castiglioni Matteo, Celli Andrea, Gatti Nicola
    34th AAAI Conference on Artificial Intelligence, AAAI 2020, New York, USA
    [paper]
  • Be a Leader or Become a Follower: The Strategy to Commit to with Multiple Leaders
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    28th International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China
    [paper]
  • Leadership in Congestion Games: Multiple User Classes and Non-Singleton Actions
    Marchesi Alberto, Castiglioni Matteo, Gatti Nicola
    28th International Joint Conference on Artificial Intelligence, IJCAI 2019, Macao, China
    [paper]
  • Journal Papers
  • Maximizing Revenue from Selfish Agents in Crowd Tasks: Indirect Incentive Strategies
    Montazeri Mina, Castiglioni Matteo, Romano Giulia, Hamed Kebriaei, Gatti Nicola
    IEEE Control Systems Letters, 2024
  • Increasing Revenue in Bayesian Posted Price Auctions through Signaling
    Castiglioni Matteo, Marchesi Alberto, Romano Giulia, Gatti Nicola
    Artificial Intelligence, 2023
    [paper]
  • Public Bayesian Persuasion: Being Almost Optimal and Almost Persuasive
    Castiglioni Matteo, Celli Andrea, Gatti Nicola
    Algorithmica, 2023
    [paper][arXiv]
  • Designing Menus of Contracts Efficiently: the Power of Randomization
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    Artificial Intelligence, 2023
    [paper]
  • Regret minimization in online Bayesian persuasion: Handling adversarial receiver's types under full and partial feedback models
    Castiglioni Matteo, Celli Andrea, Marchesi Alberto, Gatti Nicola
    Artificial Intelligence, 2023
    [paper]
  • A framework for safe decision making: A convex duality approach
    Bernasconi Martino, Cacciamani Federico, Castiglioni Matteo
    Intelligenza Artificiale, 2023
    [paper]
  • Bayesian Agency: Linear Versus Tractable Contracts
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    Artificial Intelligence, 2022
    [paper]
  • Election Manipulation on Social Networks: Seeding, Edge Removal, Edge Addition
    Castiglioni Matteo, Ferraioli Diodato, Gatti Nicola, Landriani Giulia
    Journal of Artificial Intelligence Research, 2021
    [paper]
  • Committing to Correlated Strategies with Multiple Leaders
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola
    Artificial Intelligence, 2021
    [paper]
  • Leadership in singleton congestion games: What is hard and what is easy
    Castiglioni Matteo, Marchesi Alberto, Gatti Nicola, Coniglio Stefano
    Artificial Intelligence, 2019
    [paper]
  • Teaching

  • Online Learning Applications
    M.Sc. in Computer Science and Engineering
    Professor during the academic years 2022-2023, 2023-2024, 2024-2025
  • Informatica A
    B.Sc. in Engineering Physics
    Professor during the academic years 2024-2025
  • Game Theory
    M.Sc. in Mathematical Engineering
    TA during the academic year 2020-2021, 2021-2022
  • Game Theory
    M.Sc. in Computer Science and Engineering
    TA during the academic year 2019-2020