Learning to select computations

Frederick Callaway, Sayan Gul, Paul M. Krueger, Thomas L. Griffiths, Falk Lieder
Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence, PMLR R16:775-784, 2018.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR16-callaway18a, title = {Learning to select computations}, author = {Callaway, Frederick and Gul, Sayan and Krueger, Paul M. and Griffiths, Thomas L. and Lieder, Falk}, booktitle = {Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence}, pages = {775--784}, year = {2018}, editor = {Globerson, Amir and Silva, Ricardo}, volume = {R16}, series = {Proceedings of Machine Learning Research}, month = {06--10 Aug}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r16/main/assets/callaway18a/callaway18a.pdf}, url = {https://proceedings.mlr.press/r16/callaway18a.html}, abstract = {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.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Learning to select computations %A Frederick Callaway %A Sayan Gul %A Paul M. Krueger %A Thomas L. Griffiths %A Falk Lieder %B Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2018 %E Amir Globerson %E Ricardo Silva %F pmlr-vR16-callaway18a %I PMLR %P 775--784 %U https://proceedings.mlr.press/r16/callaway18a.html %V R16 %X 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. %Z Reissued by PMLR on 04 October 2026.
APA
Callaway, F., Gul, S., Krueger, P.M., Griffiths, T.L. & Lieder, F.. (2018). Learning to select computations. Proceedings of the 34th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R16:775-784 Available from https://proceedings.mlr.press/r16/callaway18a.html. Reissued by PMLR on 04 October 2026.

Related Material