Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

window-based forgetting

Window-based forgetting is a data adaptation technique in machine learning and stream processing where an algorithm retains and trains on only a bounded subset of the most recent data instances while discarding older data that falls outside this window. By maintaining a sliding or adaptive window over incoming sequential data, the system actively forgets obsolete historical information to prevent outdated patterns from biasing current models. This approach is widely used in dynamic, non-stationary environments to handle concept drift, allowing learning systems to adapt to evolving data distributions while bounding computational and memory requirements.

1 item

Learning in the Presence of Concept Drift and Hidden Contexts

Learning in the Presence of Concept Drift and Hidden Contexts

G. Widmer, M. Kubát

OrganizationsAustrian Research Institute for Artificial IntelligenceUniversity of OttawaUniversity of Vienna

Why you should read this

Presents the FLORA framework of incremental learning algorithms that dynamically adjust sample windows and reuse past concept descriptions to handle recurring hidden contexts and concept drift in continuous data streams.

On-line learning in domains where the target concept depends on some hidden context poses serious problems. A changing context can induce changes in the target concepts, producing what is known as concept drift. We describe a family of learning algorithms that flexibly react to concept drift and can take advantage of situations where contexts reappear. The general approach underlying all these algorithms consists of (1) keeping only a window of currently trusted examples and hypotheses; (2) storing concept descriptions and re-using them when a previous context re-appears; and (3) controlling both of these functions by a heuristic that constantly monitors the system's behavior. The paper reports on experiments that test the systems' performance under various conditions such as different levels of noise and different extent and rate of concept drift.

Added

2026-09-24