论文信息 - Betweenness Centrality as a Basis for Forming Skills Özgür Şimşek

Betweenness Centrality as a Basis for Forming Skills Özgür Şimşek

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.

Andrew G. Barto | A. Barto

[1] Shie Mannor,et al. Dynamic abstraction in reinforcement learning via clustering , 2004, ICML.

[2] Leonard M. Freeman,et al. A set of measures of centrality based upon betweenness , 1977 .

[3] Thomas G. Dietterich. Hierarchical Reinforcement Learning with the MAXQ Value Function Decomposition , 1999, J. Artif. Intell. Res..

[4] Alicia P. Wolfe,et al. Local Graph Partitioning as a Basis for Generating Temporally-Extended Actions in Reinforcement Learning , 2005 .

[5] Andrew G. Barto,et al. Automatic Discovery of Subgoals in Reinforcement Learning using Diverse Density , 2001, ICML.

[6] Bernhard Hengst,et al. Discovering Hierarchy in Reinforcement Learning with HEXQ , 2002, ICML.

[7] Nuttapong Chentanez,et al. Intrinsically Motivated Learning of Hierarchical Collections of Skills , 2004 .

[8] Nuttapong Chentanez,et al. Intrinsically Motivated Reinforcement Learning , 2004, NIPS.

[9] Doina Precup,et al. Temporal abstraction in reinforcement learning , 2000, ICML 2000.

[10] L. Freeman. Centrality in social networks conceptual clarification , 1978 .

[11] Shie Mannor,et al. Q-Cut - Dynamic Discovery of Sub-goals in Reinforcement Learning , 2002, ECML.

[12] Doina Precup,et al. Between MDPs and Semi-MDPs: A Framework for Temporal Abstraction in Reinforcement Learning , 1999, Artif. Intell..