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

Download:

Current browse context:

cs.RO

Change to browse by:

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 > Robotics

Title: Q-Search Trees: An Information-Theoretic Approach Towards Hierarchical Abstractions for Agents with Computational Limitations

Abstract: In this paper, we develop a framework to obtain graph abstractions for decision-making by an agent where the abstractions emerge as a function of the agent's limited computational resources. We discuss the connection of the proposed approach with information-theoretic signal compression, and formulate a novel optimization problem to obtain tree-based abstractions as a function of the agent's computational resources. The structural properties of the new problem are discussed in detail, and two algorithmic approaches are proposed to obtain solutions to this optimization problem. We discuss the quality of, and prove relationships between, solutions obtained by the two proposed algorithms. The framework is demonstrated to generate a hierarchy of abstractions for a non-trivial environment.
Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI); Information Theory (cs.IT); Machine Learning (stat.ML)
Journal reference: 2020 IEEE Transactions on Robotics (T-RO)
DOI: 10.1109/TRO.2020.3003219
Cite as: arXiv:1910.00063 [cs.RO]
  (or arXiv:1910.00063v1 [cs.RO] for this version)

Submission history

From: Daniel Larsson [view email]
[v1] Mon, 30 Sep 2019 19:21:41 GMT (1206kb,D)

Link back to: arXiv, form interface, contact.