"The pursuit of science seems to me to
require particular courage.
It is concerned with knowledge, achieved through doubt.
Making knowledge about everything available for everybody,
science strives to make sceptics of them all."
Bertolt Brecht, Life of Galileo.
"Non credo che la pratica della scienza possa andar
disgiunta dal coraggio.
Essa tratta il sapere, che è un prodotto
del dubbio; e col procacciare sapere a tutti su ogni cosa, tende a destare
il dubbio in tutti."
Bertolt Brecht, Vita di Galileo.
Alessio Micheli
Universita` di Pisa - Dipartimento di Informatica
Largo B. Pontecorvo 3, 56127 Pisa, ITALY
Phone: +39-050-2212798 - Fax: +39-050-2212726
Room 358 / DN
Short CV
Research Interests
- Machine Learning, Artificial and Computational Intelligence, Soft Computing
- Learning in Structured Domains, Relational Learning, Relational Data Mining
- Graph Representation Learning , Learning on Graphs, Deep Graph Networks, Neural Networks for Graphs, Graph Neural Networks
- Supervised and Unsupervised Neural Networks, Recurrent and Recursive Neural Networks
- Deep Learning, Deep Recurrent Networks, Deep Reservoir Computing, DeepESN
- Processing of Sequences and Structured Information (trees and graphs) in Machine Learning:
Recursive Models, Reservoir Computing, Hidden Markov Models, Kernel-based methods for Non-vectorial Data
- Applications to Cheminformatics, QSPR/QSAR Analysis, and Toxicity predictions
- Applications to Bioinformatics and Health/biomedical Informatics
- Learning in Robotics and Wireless/Intelligent Sensor Networks
- Human Activity Recognition for Ambient Assisted Living
- Pioneering since the 90’s the development of new efficient models for learning structured and graph data
(see also Methodologies and Models in CIML for
Recursive and Contextual Neural Networks topic)
- introducing the first (constructive) model of Neural Network for Graphs in the class of deep convolution approaches [2005-2009] (see IEEE TNN 2009 and the references therein)
- introducing a deep architecture for the Reservoir Computing for sequences (Neurocomputing 2017, Neural Networks 2018),
trees (Information Science 2019), and graphs ( AAAI 2020)
- providing a survey for the Deep Graph Networks (Neural Networks 2020)
- introducing DynGESN, a deep reservoir computing model for discrete-time dynamic graphs (Neurocomputing 2022)
Leading since 2008 the CIML
group for basic and applied research on Machine Learning for complex data.
Coordinations/Memberships
Research Group
- I’m proud to lead the CIML (Computational Intelligence and Machine Learning) research group, with more than 20 years of basic and applied research in Machine Learning
I am fortunate to supervise and to have supervised several excellent PhD students:
Claudio Gallicchio,
Rita Pucci,
Luca Pedrelli, Daniele Di Sarli,
Marco Podda,
Federico Errica,
Domenico Tortorella,
Marco Cardia, Sara Joubbi, Michele Fontanesi.
I am enthusiastic to work with several postdocs and researchers, see the CIML site: https://ciml.di.unipi.it/
Editorial and Program Committees
Editorial Committes
PC Conferences 2013-2022 (selection):
- IJCAI 2024 Senior program committee
- IJCAI-ECAI 2022 Senior program committee
- IJCAI (International Joint Conference on Artificial Intelligence)
- ECAI (European Conference on Artificial Intelligence)
- ICANN (International Conference on Artificial Neural Networks) Steering Committee since 2020
- IEEE SCCI CIDM (IEEE Symposium on Computational Intelligence and Data Mining)
- IEEE SCCI DL (IEEE Symposium on Deep Learning)
- IEEE IJCNN (International Joint Conference on Neural Networks)
- ESANN (European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning)
- IEEE INISTA (IEEE International Symposium on INnovations in Intelligent SysTems and Applications)
- ICPRAM (International Conference on Pattern Recognition Applications and Methods)
- IEEE ETFA (Emerging Technologies and Factory Automation, Track 10 PC)
- NC2 (Workshop on New Challenges in Neural Computation and Machine Learning)
- MOD (International Workshop on Machine Learning, Optimization and Big Data)
- IML (International Conference on Internet of Things and Machine Learning)
- AIxIA (International Conference of the Italian Association for Artificial Intelligence)
- WIRN (Italian Workshop on Neural Networks)
- Also involved as reviewer for NeurIPS 2019 (best reviewers list) & 2020, AAAI 2020, ICML 2021
Organizer of the
Publications
See also (for full references):
Teaching
- Sono disponibili Proposte di Tesi sul Machine Learning: contattare all'indirizzo micheli@di.unipi.it, tag: [Richiesta-Tesi]
Projects
Current Projects
- NRRP Extended Partnerships "Artificial Intelligence: Fundamental aspects": FAIR - Future Artificial Intelligence Research, NextGeneration EU programme, 3 years since January 2023
- NRRP Ecosystems: "THE - Tuscany Health Ecosystem", NextGeneration EU programme, 3 years since December 2022
- EU H2020 TAILOR: Foundations of Trustworthy AI - Integrating Reasoning, Learning and Optimization [2020-2023]
- EU H2020 TEACHING : A computing toolkit for building efficient autonomous applications leveraging humanistic intelligence [2020-2023]
- BrAID: Brugada syndrome and Artificial Intelligence applications to Diagnosis (Bando Ricerca e Salute 2018 - Regione Toscana) [2020-2024] (Unit leader)
- ADDSTRES: Data-Driven Approaches for Resilient and Sustainable Transport Services (Regione Toscana) [2022-2024]
Recent Projects
- EU FP7 RUBICON: Robotics UBIquitous COgnitive Network (WP leader)
- EU FP7 DOREMI:
Decrease of cOgnitive decline, malnutRition and sedEntariness by elderly
empowerment in lifestyle Management and social Inclusion (WP leader)
- ASPIS: Early seismic alert for sensitive infrastructures (POR FSE 2014-2020) (Unit leader)
Current/Recent University-Industry Collaborations
- FOSBER S.p.A: Machine Learning applied to corrugated cardboard production (2021-2022)
- MAIOR: Data-Driven Approaches for Resilient and Sustainable Transport Services (2022-2023)
- TAGES: Artificial Intelligence for mobility monitoring and management (2019)
Past Projects
- International Project: "Adaptive Processing of Data Structures" (Australian Research Council)
- National Project: "Design of biologically active molecules by neural network techniques applied to QSAR studies and by investigation of ligand-macromolecule interactions".
("Progettazione di molecole biologicamente attive mediante studi QSAR con tecniche neurali e studi di interazione ligando-macromolecola") ( Cofin 99 - MURST and University of Pisa)
- National Projects PRIN - MIUR: Cofin 1999, Cofin 2000, Cofin 2001, Cofin 2002, Cofin 2003, Cofin 2005.
E-mail: micheli@di.unipi.it