A multi-label Continual Learning framework to scale deep learning approaches for packaging equipment monitoring

Davide Dalle Pezze, Denis Deronjic, Chiara Masiero, Diego Tosato, Alessandro Beghi, Gian Antonio Susto

Abstract: Highlights•We investigate the scenario where a model adapts to handle a stream of machines with distribution shifts.•Tests on real packaging data proved the feasibility of Continual Learning for addressing such problems.•The study uncovers the limitations of previous algorithms in the Domain Incremental Learning.•We introduce a novel approach that surpasses existing literature methods in terms of performance.•Our method also has logarithmic complexity, significantly reducing computation times.