Year: 2016
Type: conference-paper
Abstract: Recently, incremental and on-line learning gained more attention \nespecially in the context of big data and learning from data streams,\nconflicting with the traditional assumption of complete data availability. Even though \na variety of different methods are available, it often remains unclear which of them is suitable\nfor a specific task and how they perform in comparison to each other. \nWe analyze the key properties of seven incremental methods representing different algorithm classes.\nOur extensive evaluation on data sets with different characteristics gives an overview of the performance with respect to accuracy as well as model complexity,\nfacilitating the choice of the best method for a given application.
Authors: Viktor Losing, Barbara Hammer, Heiko Wersing
Institutions: Bielefeld University, Honda (Japan)
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2603537372