We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.AI

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Artificial Intelligence

Title: Autonomous Extracting a Hierarchical Structure of Tasks in Reinforcement Learning and Multi-task Reinforcement Learning

Abstract: Reinforcement learning (RL), while often powerful, can suffer from slow learning speeds, particularly in high dimensional spaces. The autonomous decomposition of tasks and use of hierarchical methods hold the potential to significantly speed up learning in such domains. This paper proposes a novel practical method that can autonomously decompose tasks, by leveraging association rule mining, which discovers hidden relationship among entities in data mining. We introduce a novel method called ARM-HSTRL (Association Rule Mining to extract Hierarchical Structure of Tasks in Reinforcement Learning). It extracts temporal and structural relationships of sub-goals in RL, and multi-task RL. In particular,it finds sub-goals and relationship among them. It is shown the significant efficiency and performance of the proposed method in two main topics of RL.
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:1709.04579 [cs.AI]
  (or arXiv:1709.04579v2 [cs.AI] for this version)

Submission history

From: Behzad Ghazanfari [view email]
[v1] Thu, 14 Sep 2017 01:43:13 GMT (316kb,D)
[v2] Fri, 15 Sep 2017 16:21:03 GMT (316kb,D)

Link back to: arXiv, form interface, contact.