Skill characterization based on betweenness

Özgür Şimşek, Andrew G. Barto

Research output: Chapter in Book/Report/Conference proceedingChapter

68 Citations (Scopus)

Abstract

We present a characterization of a useful class of skills based on a graphical representation of an agent's interaction with its environment. Our characterization uses betweenness, a measure of centrality on graphs. It may be used directly to form a set of skills suitable for a given environment. More importantly, it serves as a useful guide for developing online, incremental skill discovery algorithms that do not rely on knowing or representing the environment graph in its entirety.
Original languageEnglish
Title of host publicationAdvances in Neural Information Processing Systems 21 (NIPS 2008): Proceedings of the Twenty-Second Annual Conference on Neural Information Processing Systems, Vancouver, British Columbia, Canada, December 8-11, 2008
EditorsDaphne Koller, Dale Schuurmans, Yoshua Bengio, Léon Bottou
PublisherCurran Associates, Inc.
Pages1497-1504
Number of pages8
Publication statusPublished - 2009

Cite this

Şimşek, Ö., & Barto, A. G. (2009). Skill characterization based on betweenness. In D. Koller, D. Schuurmans, Y. Bengio, & L. Bottou (Eds.), Advances in Neural Information Processing Systems 21 (NIPS 2008): Proceedings of the Twenty-Second Annual Conference on Neural Information Processing Systems, Vancouver, British Columbia, Canada, December 8-11, 2008 (pp. 1497-1504). Curran Associates, Inc..