Header logo is

Learning to select computations

2018

Conference Paper

re


The efficient use of limited computational resources is an essential ingredient of intelligence. Selecting computations optimally according to rational metareasoning would achieve this, but this is computationally intractable. Inspired by psychology and neuroscience, we propose the first concrete and domain-general learning algorithm for approximating the optimal selection of computations: Bayesian metalevel policy search (BMPS). We derive this general, sample-efficient search algorithm for a computation-selecting metalevel policy based on the insight that the value of information lies between the myopic value of information and the value of perfect information. We evaluate BMPS on three increasingly difficult metareasoning problems: when to terminate computation, how to allocate computation between competing options, and planning. Across all three domains, BMPS achieved near-optimal performance and compared favorably to previously proposed metareasoning heuristics. Finally, we demonstrate the practical utility of BMPS in an emergency management scenario, even accounting for the overhead of metareasoning.

Author(s): Frederick Callaway and Sayan Gul and Paul M. Krueger and Thomas L. Griffiths and Falk Lieder
Book Title: Uncertainty in Artificial Intelligence: Proceedings of the Thirty-Fourth Conference
Year: 2018
Month: August

Department(s): Rationality Enhancement
Research Project(s): Metacognitive Learning
Bibtex Type: Conference Paper (inproceedings)

Note: Frederick Callaway and Sayan Gul and Falk Lieder contributed equally to this publication.
State: Published
URL: https://arxiv.org/pdf/1711.06892.pdf
Attachments:

BibTex

@inproceedings{Callaway2018Learning,
  title = {Learning to select computations},
  author = {Callaway, Frederick and Gul, Sayan and Krueger, Paul M. and Griffiths, Thomas L. and Lieder, Falk},
  booktitle = {{Uncertainty in Artificial Intelligence: Proceedings of the Thirty-Fourth Conference}},
  month = aug,
  year = {2018},
  note = {Frederick Callaway and Sayan Gul  and Falk Lieder contributed equally to this publication.},
  url = {https://arxiv.org/pdf/1711.06892.pdf},
  month_numeric = {8}
}