🚀
Research
Where you can find my Research Interests and Activities
I have a strong background in Parallel Computing and Data Analytics. My interest has then evolved towards the field of Artificial Intelligence, having always been intrigued by the idea of building machines as intelligent as humans.
I am now fully engaged in the study of Deep Learning, Continual / Lifelong Learning, Knowledge Transfer and Distillation, Distributed Learning and their applications.
The long-term goal of my research aims at answering this single question:
What could make artificial intelligence systems as sustainable (efficient & scalable) and effective as the human brain? And how to implement it algorithmically?
While the answer to the “what” part of the question is obvious to me (at least on where to start😝) and it is Continual Learning, the “how” part is the main focus of my research. This mostly means understanding the true nature of biological learning systems that learn continually: by what computational principles they are governed and how the interaction of these principles let to the emergence of intelligent behaviors; and secondly, the design, prototype and experimentation of algorithmic formulations for artificial learning systems.
My research in Continual Learning is split in three main focus areas: Continual Learning in Deep Learning (and Neuroscience-Inspired AI), Continual Learning and its relationships with Distributed/Federated Learning and Continual Learning for Practical Applications, all within a Sustainable AI developmental framework.
These four different focus areas ensure that my research in Continual Learning is grounded in (Deep) Machine Learning, while also preserving a certain degree of (sustainable) practical applicability. This is important to evaluate the reachable level of robustness and maturity of our algorithmic proposals, gather new insights, making a tangible impact in the real-world as well as making sure to direct and take responsibility for this change.
Assistant Professor (Junior)
2021 - Present Department of Computer Science and Engineering, University of Pisa
Research topics: Artificial Intelligence, Machine Learning, Continual Learning.
EMERGE: Emergent Awareness from Minimal Collectives
2022 - Present
EU-funded EMERGE project which will enable emergent collaborative awareness among collectives of minimal artificial beings without predefined protocols. It will support complex, distributed and flexible systems exhibiting collaboration, self-regulation and interoperability. EMERGE’s framework will offer insight that will foster allocation of awareness to optimise achievement of desired outcomes.
Involvement: Taks Leader
Continual Learning for Computer Vision
2022 - Present
Research contract with the Leonardo S.p.a. for the research and development of original continual learning solutions applied to Computer Vision.
Involvement: Principal Investigator
Continual Learning for Predictive Maintenance
2021 - Present
Research contract with the SeaVision Group for the research and development of original continual learning solutions applied to predictive maintenance in the pharmaceutical domain.
Involvement: Principal Investigator
TEACHING: A computing Toolkit for building Efficient Autonomous appliCations leveraging Humanistic INtelliGence
2021 - Present
EU-funded project that designs a computing platform and the associated software toolkit supporting the development and deployment of autonomous, adaptive and dependable CPSoS applications, allowing them to exploit a sustainable human feedback to drive, optimize and personalize the provisioning of their services.
Involvement: Task leader
Continual Learning for Object Detection
2022 - 2022
Gift contract with the Meta for the research and development of object detection support in Avalanche for the development of the CLVision 2022 Workshop challenge competition.
Involvement: Principal Investigator
Continual Learning in Computer Vision
2021 - 2021
Gift contract with the Intel for the support of the CLVision 2020 Workshop challenge competition.
Involvement: Principal Investigator
Image Manipulation Attack Resolving Solutions
2020 - 2021
The iMARS research project, fully funded by the European Union for a total budget of 7+ Million, aims at designing new solutions for the detection of morphing attacks and digital image manipulations in order to provide accurate verification of ID documents.
Involvement: Participant
Recycle Symbols Detection on Embedded Cameras
2020 - 2021
In collaboration with the Hera Group, one of the biggest multi-services companies in Italy, the project was centered around the prototyping of an object detection API service running directly on the edge on Android mobile phones.
Involvement: Participant
General Chair / Main Organizer
2021 - Present
Deep Continual Learning, Dagstuhl Seminar 23122, March 19 – 24 , 2023 (Participants: 50, invite-only)
1st ed. of the Workshop "Novel Benchmarks and Approaches for Real-World Continual Learning"
ICIAP 2021. (Participants: 100-200)
3nd ed. of the Workshop "Continual Learning for Computer Vision"
CVPR 2022. (Participants: 300-400)
1st ed of the Workshop "Semi-Supervised Continual Learning Workshop"
IJCAI 2021. (Participants: 250-300)
1st ed. of the Workshop "Theory of Continual Learning Workshop"
ICML 2021. (Participants: 250-300)
2nd ed. of the Workshop "Continual Learning for Computer Vision"
CVPR 2021. (Participants: 250-300)
1st ed. of the Workshop "Continual Learning for Computer Vision"
CVPR 2020. (Participants: 250-300)
Main organizer of the ContinualAI Monthly meetups and weekly seminars
ContinualAI, 2021 - Present. (Participants: 30-60)
Program Chair
2021 - Present
"AI for People" conference 2021, Online.
Special Session Chair
2021 - Present
"Continual Learning and Emergence of Intelligent Systems: Theory and Application" (IJCNN 2022)
"Advances in Continual Learning: beyond Catastrophic Forgetting" (IEEE EAIS 2021).
"AI for People" (GOODTECHS 2020).
"New Trends in Continual Learning with Deep Architectures" (IEEE EAIS 2020).
Program Committee Member & Reviewer
2016 - Present
IJCAI-PRICAI 2020, IEEE EAIS 2020, GOODTECHS 2020, Continual Learning workshop at ICML 2020, ROMAN Workshop on Lifelong Learning for Long-term HRI, CCNC 2021 Workshop RoboCom, AAAiI-2022, EAIS 2022, CCNC 2022 Workshop RoboCom, PeRConAI 2022, IEEE RA-L, ICPR 2020, IROS 2020, ICML 2020 Continual Learning workshop, ICONIP 2020, ICML 2020 Lifelong Machine Learning, ISBA 2016, ISBA 2017, ISBA 2018, PIMRC2018, CoRL 2017, ICANN 2019, AAAI 2020, ECAI 2020, EAIS 2020, CVPR 2022, ICIAP 2022, ECCV 2022, Conference on Lifelong Learning Agents (CoLLAs) 2022 and a reviewer for prestigious journals / institutions such as Elsevier Cognitive Systems Research, Artificial Intelligence in Medicine and IEEE Access, Journal of Information Security and Applications, Artificial Intelligence Journal, IEEE Access, CVPR 2021, Neural Computing and Applications Journal, ICRA 2021, ICCV 2021, IEEE Transactions on Neural Networks and Learning Systems, Neurocomputing, Mobile Information Systems Hindawi, Evolving Systems, Frontiers Robotics and AI, Elsevier Patter Recognition, TPAMI, Leverhulme Trust, Nature Machine Intelligence among others.
Author, Editor and Technical Reviewer
2016 - Present
"Continual Semi-Supervised Learning". Fabio Cuzzolin, Kevin Cannons, Vincenzo Lomonaco. Springer 2022. (Book chapter Editor)
"Deep Learning with R". Vincenzo Lomonaco. Packt Publishing, 2017 (Author).
"Online continual learning on sequences." Parisi, German I., and Vincenzo Lomonaco in "Recent Trends in Learning From Data". Oneto, L., Navarin, N., Sperduti, A., Anguita, D., Springer, 2019. (Book Chapter Author)
"Intelligenza Artificiale", IOS Press, 2022-Present. (Associate Editor)
"R Deep Learning Essentials. Packt Publishing". Joshua F. Wiley, 2016. (Technical Reviewer)
Special Issue Editor
2020 - Present
"Continual Unsupervised Learning in Computer Vision", Frontiers in Computer Science, 2022.
"Adaptive Machines: Leveraging Neuroscience for Lifelong Learning Systems" for the Frontiers in Artificial Intelligence Journal, 2021.
"Lifelong Learning Machines" for the Elsevier Neural Networks Journal, 2021.
"AI for People" for the Springer Journal "AI & Society: Journal of Knowledge, Culture and Communication", 2020.
Co-Founder & President
2018 - Present
President of the non-profit research organization ContinualAI, 2018 - Present
ContinualAI is the largest research organization and open community on Continual Learning for Artificial Intelligence. The organization has more than 1000+ members in 19 different timezones, 5000+ annual users of its online services, 30+ supporting partners, 50+ organized or supported events, 5 active collaborative projects, e 3000+ followers on its social platforms.
Website: https://www.continualai.org
Co-Founder & Board Director of AIforPeople, 2019 - Present
Co-Founder & Board Member of the non-profit organization AIforPeople. The social mission of AIforPeople is the one to learn, pose questions and take initiative on how the technology based on Artificial Intelligence can be used for the social good.
Website: https://www.aiforpeople.org
International Doctoral Reviewer
2020 - Present
"Tackling the Distribution Shift in Visual Understanding Applications", Marco Toldo, Padova University, 2022.
"Dissecting Continual Learning a Structural and Data Analysis”, Francesco Pelosin, Ca' Foscari University, 2022.
"Continual Object Detection with Deep Neural Networks", Ângelo Garangau Menezes, Universidade de São Paulo, 2022.
"Continual learning for hierarchical classification, meta-learning, and multi-modal learning", Kai Wang, Universitat Autonoma de Barcelona, 2022.
"Augmenting Deep Learning models using Continual and Meta Learning strategies", Julio Hurtado, Pontificia Universidad Católica de Chile, 2022.
"Neural Architecture Search under Budget Constraints", Tom Veniat, Sorbonne Université, 2021.
"Lifelong Learning of Neural Networks: Detecting Novelty and Adapting to New Domains without Forgetting", Marc Masana, Universitat Autonoma de Barcelona, 2020.
Associations
2018 - Present
ELLIS, CLAIR, IEEE, IEEE Computational Intelligence Society, AIXIA, CVPL, IAML, AIforPeople, ContinualAI, CINI AIIS.
Awards
2009 - Present
W&B Best Library Award: Avalanche was chosen by Weights & Bias as the best Continual Learning library presented ad CLVision 2021.
Doctoral Dissertation Distinction: The dissertation was awarded by the Italian Association for Artificial Intelligence as one of the top-5 AI PhD Thesis of 2019.
Challenge Winner: 2nd classified of the competition "Lifelong Robotic Vision" organized at IROS 2019 with the UniBo team supervised by Prof. Davide Maltoni.
Hackaton Winner: 1st classified at the Hackathon "HackCortona" (KCL Tech + Cortona Mix Festival). Cortona, Italy, 2016.
Best Short-Film Award: winner of the National Short-Film Festival "L'educazione fa crescere i diritti" organized by CISP, Roma, Italy, 2009.
Certificates
2015 - Present
English Proficiency Certificate IELTS: "International English Language Testing System" certificate. Grade 7, 2015.
Podcasting Activity
2021 - Present
Smarter Podcast: Il Podcast Italiano sull' Intelligenza Artificiale, Co-Organizer and Co-Host, 2021 - Present.
TEACHING Podcast: Official podcast of the EU Funded Project TEACHING, A computing Toolkit for building Efficient Autonomous appliCations leveraging Humanistic INtelliGence, Lead Organizer and Host, 2021 - Present.
Pointer Podcast: Guest Interview on Continual Learning, Avalanche and more, Episode 99th, 4th May, 2022.
Press Articles
2021 - Present
AI Weekly: Continual learning offers a path toward more humanlike AI, Interview, VentureBeat, 2021.
L’IA attuale non è sostenibile: cosa cambia con l’apprendimento automatico continuo, Author, Agenda Digitale, 2022.
Università di Pisa, al via il progetto europeo per una Intelligenza artificiale, Sole 24 ore, 2022.
Dissemination Videos
2016 - Present
Tutorials
2019 - Present
Continual Lifelong Learning with Neural Networks, G.I. Parisi and V. Lomonaco, INNSBDDL 2019. [Slides]
Scientific Talks
2016 - Present
Introduction to Continual Learning, Advanced Topics in Machine Learning, Università della Svizzera Italiana, Lugano, 22-11-22. [Slides]
Introduction to Continual Learning, Advances in AI, International Summer School, 10-09-22. [Slides]
Fundamentals of Continual Learning and Continual Learning in Practice, Summer School Lecture, Summer School of Information Engineering (SSIE), 12-07-22. [Slides 1] [Slides 2]
Introduction to Continual Learning, Continually Learning Biometrics, Summer School Lecture, 07-06-2022. [Slides]
Introduction to Continual Learning, Intelligent Systems and Pattern Recognition course, University of Pisa, 20-04-22. [Slides]
Ex-Model Continual Learning: a New Paradigm for Distributed Robotics Intelligence, Invited Talk @ Workshop on Lifelong Learning and Personalization in Long-Term Human-Robot Interaction 2022. [Video] [Slides]
Introduction to Continual Learning, AI Seminars, Sapienza University, Rome, Italy, 24-3-22. [Slides]
Introduction to Continual Learning, School of AI, University of Modena and Reggio-Emilia, 2022. [Slides]
Distributed Continual Learning: Challenges and Opportunities, Invited Talk, Workshop on Continual Learning and Adaptation for Time Evolving Data, 2021. [Video] [Slides]
Sustainable AI through Continual Learning, SMART Cloud #3 - AI & Machine Learning, 10-19-2021. [Slides].
Introduction to Continual Learning, Intelligent Systems & Pattern Recognition course, University of Pisa, 20-03-21. [Slides]
The Thousand Brains Theory of Intelligence, Computational Neuroscience Course, University of Pisa, 26-05-2021. [Slides]
CORe: an Android App for Continual Object Recognition at the Edge, Workshop on On-Device Machine Learning, 4-04-2021. [Slides]
Sustainable AI through Continual Learning, Department of Computer Science - University of Pisa 22-03-2021. [Slides]
Rehearsal-Free Continual Learning over Small non-I.I.D Batches, ContinualAI Meetup, 27-11-2020. [Slides]
Sustainable AI through Continual Learning, Continual Learning: Towards “Broad” AI course, Université de Montréal, 25-01-2021.
Open-Source Frameworks for Deep Learning: an Overview, Machine Learning Course, University of Bologna, Italy, 2020. [Slides]
Real-Time Continual Learning from Natural Video Streams, Human-centered Vision: from Body Analysis to Learning and Language Workshop, Online, 9 September 2020.
Continual Learning, Neuroscience and Robotics: an Entwined Destiny, Invited talk @ Lifelong Learning for Long-term Human-Robot Interaction, ROMAN2020, 04-09-2020. [Video] [Slides]
Towards Continual Learning at the Edge Talk, HiPeRT Lab, University of Modena and Reggio-Emilia. 18-05-2020. [Slides]
New Trends in Continual Learning with Deep Architectures, Special Session @ IEEE EAIS2020 27-05-2020. [Slides]
Continual Learning for Production Systems, Talk @ PRODUCTION.AI conference, Kiev, Ukraine, 27-05-2020. [Slides]
Open-Source Frameworks for Deep Learning: an Overview, Machine Learning Course, University of Bologna, 2019. [Slides]
Continual Learning: Another Step Towards Truly Intelligent Machines, Research Meeting @ Numenta, California, USA, 23-09-2019. [Slides]
Continual Learning for Robotics, Italian Institute of Technology, Genova, Italy, 11-09-2019. [Slides]
Apprendimento Automatico Continuo per la Robotica e l'Intelligenza Artificiale, Codemotion Meetup @ AlmaCube, Bologna, Italy, 4-07-2019. [Slides]
Continual Reinforcement Learning in 3D Non-stationary Environments, UPF - Computational Science Lab, Barcelona, Spain, 29-03-2019. [Slides]
Continual Learning with Deep Architectures, Workshop @ Data Science Milan, Milan, Italy, 28-01-2019. [Slides]
Continual Learning for Robotics and Computer Vision, Ital-IA conference, Rome, Italy, 18-03-2019. [Slides]
Open-Source Frameworks for Deep Learning: an Overview, Musixmatch, Bologna, Italy 28-02-2018. [Slides]
Continual Learning with Deep Architectures, Workshop @ Computer VISIONers Conference, 06-10-2018, Kiev, Ukraine. [Slides]
CORe50: A new Dataset and Benchmark for Continuous Object Recognition, CoRL 2017, Google HQ, Mountain View, USA. [Article] [Video] [Slides]
Last modified 11d ago