Betweenness Centrality as a Basis for Forming Skills

Özgür Şimşek; Andrew G. Barto

Betweenness Centrality as a Basis for Forming Skills

Özgür Şimşek, Andrew G. Barto

University of Massachusetts

Research output: Working paper / Preprint › Working paper

71 Downloads (Pure)

Abstract

We show that betweenness centrality, a graph-theoretic measure widely used in social network analysis, provides a sound basis for autonomously forming useful high-level behaviors, or skills, from available primitives— the smallest behavioral units available to an autonomous agent.

Original language	English
Publisher	University of Massachusetts Amherst
Publication status	Published - 12 Apr 2007

Keywords

reinforcement learning
skill discovery
Hierarchical reinforcement learning
action hierachy

Access to Document

simsek_b_TR07-26Final published version, 805 KB

Cite this

@techreport{a2bf93d1690c478fb5f6dc7cf752cbb6,

title = "Betweenness Centrality as a Basis for Forming Skills",

abstract = "We show that betweenness centrality, a graph-theoretic measure widely used in social network analysis, provides a sound basis for autonomously forming useful high-level behaviors, or skills, from available primitives— the smallest behavioral units available to an autonomous agent.",

keywords = "reinforcement learning, skill discovery, Hierarchical reinforcement learning, action hierachy",

author = "{\"O}zg{\"u}r {\c S}im{\c s}ek and Barto, \{Andrew G.\}",

year = "2007",

month = apr,

day = "12",

language = "English",

publisher = "University of Massachusetts Amherst",

type = "WorkingPaper",

institution = "University of Massachusetts Amherst",

}

TY - UNPB

T1 - Betweenness Centrality as a Basis for Forming Skills

AU - Şimşek, Özgür

AU - Barto, Andrew G.

PY - 2007/4/12

Y1 - 2007/4/12

N2 - We show that betweenness centrality, a graph-theoretic measure widely used in social network analysis, provides a sound basis for autonomously forming useful high-level behaviors, or skills, from available primitives— the smallest behavioral units available to an autonomous agent.

AB - We show that betweenness centrality, a graph-theoretic measure widely used in social network analysis, provides a sound basis for autonomously forming useful high-level behaviors, or skills, from available primitives— the smallest behavioral units available to an autonomous agent.

KW - reinforcement learning

KW - skill discovery

KW - Hierarchical reinforcement learning

KW - action hierachy

M3 - Working paper

BT - Betweenness Centrality as a Basis for Forming Skills

PB - University of Massachusetts Amherst

ER -

Betweenness Centrality as a Basis for Forming Skills

Abstract

Keywords

Access to Document

Fingerprint

Cite this