Classification in non-metric spaces

Daphna Weinshall, David W. Jacobs, Yoram Gdalyahu

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

18 Scopus citations


A key question in vision is how to represent our knowledge of previously encountered objects to classify new ones. The answer depends on how we determine the similarity of two objects. Similarity tells us how relevant each previously seen object is in determining the category to which a new object belongs. Here a dichotomy emerges. Complex notions of similarity appear necessary for cognitive models and applications, while simple notions of similarity form a tractable basis for current computational approaches to classification. We explore the nature of this dichotomy and why it calls for new approaches to well-studied problems in learning. We begin this process by demonstrating new computational methods for supervised learning that can handle complex notions of similarity. (1) We discuss how to implement parametric methods that represent a class by its mean when using non-metric similarity functions; and (2) We review non-parametric methods that we have developed using nearest neighbor classification in non-metric spaces. Point (2), and some of the background of our work have been described in more detail in [8].

Original languageAmerican English
Title of host publicationAdvances in Neural Information Processing Systems 11 - Proceedings of the 1998 Conference, NIPS 1998
PublisherNeural information processing systems foundation
Number of pages7
ISBN (Print)0262112450, 9780262112451
StatePublished - 1999
Event12th Annual Conference on Neural Information Processing Systems, NIPS 1998 - Denver, CO, United States
Duration: 30 Nov 19985 Dec 1998

Publication series

NameAdvances in Neural Information Processing Systems
ISSN (Print)1049-5258


Conference12th Annual Conference on Neural Information Processing Systems, NIPS 1998
Country/TerritoryUnited States
CityDenver, CO


Dive into the research topics of 'Classification in non-metric spaces'. Together they form a unique fingerprint.

Cite this