Antonio Carta
I am an assistant professor at the Department of Computer Science of the University of Pisa, working with the Computational intelligence and Machine Learning group (CIML) and the Pervasive AI Lab (PAILab). I am also a member of ContinualAI, where I am the lead maintainer of Avalanche, a continual learning library.
My main research interest is continual learning with deep neural networks. I am currently working on efficient methods to share and transfer knowledge between different continual learning models (e.g. distributed continual learning models). I am also working on efficient and robust methods for online continual learning. Applications include both computer vision, natural language, and time series.
Recent Research
A Simple Recipe to Meta-Learn Forward and Backward Transfer SiM4C, a simple and robust method for meta-continual and continual-meta learning. ICCV23A Comprehensive Empirical Evaluation on Online Continual Learning Large OCL evaluation with several metrics and baselines. VCL workshop @ICCV23 (preprint)
Recent Talks
CoLLAs 2023 Tutorial "Continual Learning Beyond Catastrophic Forgetting in Class-Incremental Scenarios" by me and Vincenzo Lomonaco
(slides).