no code implementations • 8 Dec 2022 • Indranil Sur, Zachary Daniels, Abrar Rahman, Kamil Faber, Gianmarco J. Gallardo, Tyler L. Hayes, Cameron E. Taylor, Mustafa Burak Gurbuz, James Smith, Sahana Joshi, Nathalie Japkowicz, Michael Baron, Zsolt Kira, Christopher Kanan, Roberto Corizzo, Ajay Divakaran, Michael Piacentino, Jesse Hostetler, Aswin Raghavan
In this paper, we introduce the Lifelong Reinforcement Learning Components Framework (L2RLCF), which standardizes L2RL systems and assimilates different continual learning components (each addressing different aspects of the lifelong learning problem) into a unified system.
no code implementations • 18 Jan 2022 • Kamil Faber, Roberto Corizzo, Bartlomiej Sniezynski, Michael Baron, Nathalie Japkowicz
Detecting relevant changes in dynamic time series data in a timely manner is crucially important for many data analysis tasks in real-world settings.