Defying the gravity of learning curve : A characteristic of nearest neighbour anomaly detectors

Title: Defying the gravity of learning curve : A characteristic of nearest neighbour anomaly detectors
Creator: Ting, Kaiming; Washio, Takashi; Wells, Jonathan; Aryal, Sunil
Date: 2017
Type: Text; Journal article
Identifier: http://researchonline.federation.edu.au/vital/access/HandleResolver/1959.17/163303
Identifier: vital:12854
Identifier: https://doi.org/10.1007/s10994-016-5586-4
Identifier: ISBN:0885-6125
Abstract: Conventional wisdom in machine learning says that all algorithms are expected to follow the trajectory of a learning curve which is often colloquially referred to as ‘more data the better’. We call this ‘the gravity of learning curve’, and it is assumed that no learning algorithms are ‘gravity-defiant’. Contrary to the conventional wisdom, this paper provides the theoretical analysis and the empirical evidence that nearest neighbour anomaly detectors are gravity-defiant algorithms.
Relation: Machine Learning Vol. 106, no. 1 (2017), p. 55-91
Rights: This metadata is freely available under a CCO license
Subject: 0801 Artificial Intelligence and Image Processing; 1702 Cognitive Science; Learning curve; Anomaly detection; Nearest neighbour; Computational geometry; AUC
Reviewed

		Thumbnail	File	Description	Size	Format