Lifelong Continual Learning in Dynamic Environments: a Tutorial

Matteo Tiezzi gave a tutorial on Lifelong Continual Learning at the PAVIS@IIT group (Istituto Italiano di Tecnologia), Genova, Italy, where is currently a PostDoc Researcher. The tutorial gave an overview on motivations, scenarios, settings, benchmarks and the main methods of Continual Learning.

Abstract

This tutorial offers a general overview of Continual Learning (CL), a paradigm in machine learning that enables models to learn continuously from a stream of data while adapting to new tasks without forgetting previously acquired knowledge. Starting with foundational concepts and motivations, the tutorial will cover key challenges such as catastrophic forgetting and explore various CL strategies. We will discuss practical applications in real-world scenarios, highlighting the relevance of CL in dynamic environments.

Slides