Data Science Seminars - Learning From Data Streams with Streaming Continual Learning

Mercoledì 14 maggio 2025 | 17:00
Data Science and Bioinformatics Lab (Edificio 21)
Dipartimento di Elettronica, Informazione e Bioingegneria - Politecnico di Milano
Speaker: Federico Giannini (Politecnico di Milano)
Data Science and Bioinformatics Lab (Edificio 21)
Dipartimento di Elettronica, Informazione e Bioingegneria - Politecnico di Milano
Speaker: Federico Giannini (Politecnico di Milano)
Contatti: Silvia Cascianelli | silvia.cascianelli@polimi.it
Sommario
Wednesday, May 14, 2025 at 5:00 p.m. Federico Giannini (Politecnico di Milano) will hold a seminar titled "Learning From Data Streams with Streaming Continual Learning" in the Data Science and Bioinformatics Lab (Building 21). The event is part of the Data Science Seminars organized by the Data Science Lab at Politecnico di Milano.
This seminar introduces the challenges of learning from data streams, a scenario increasingly common in real-world applications where data arrives continuously and evolves over time. The discussion will explore how current research areas (Streaming Machine Learning and Continual Learning) have independently tackled these challenges, highlighting their respective settings and objectives. Building on this foundation, the seminar will present Streaming Continual Learning, a new unified framework that aims to integrate both methodologies in order to tackle all the challenges of data stream learning in a comprehensive way. The talk will also outline the principles of this manifesto and illustrate its potential through selected embodiments, including recent papers that concretely instantiate the framework.
This seminar introduces the challenges of learning from data streams, a scenario increasingly common in real-world applications where data arrives continuously and evolves over time. The discussion will explore how current research areas (Streaming Machine Learning and Continual Learning) have independently tackled these challenges, highlighting their respective settings and objectives. Building on this foundation, the seminar will present Streaming Continual Learning, a new unified framework that aims to integrate both methodologies in order to tackle all the challenges of data stream learning in a comprehensive way. The talk will also outline the principles of this manifesto and illustrate its potential through selected embodiments, including recent papers that concretely instantiate the framework.