Conformal Prediction Sets Improve Human Decision Making

Jesse C. Cresswell, Yi Sui, Bhargava Kumar, Noël Vouitsis
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:9439-9457, 2024.

Abstract

In response to everyday queries, humans explicitly signal uncertainty and offer alternative answers when they are unsure. Machine learning models that output calibrated prediction sets through conformal prediction mimic this human behaviour; larger sets signal greater uncertainty while providing alternatives. In this work, we study the usefulness of conformal prediction sets as an aid for human decision making by conducting a pre-registered randomized controlled trial with conformal prediction sets provided to human subjects. With statistical significance, we find that when humans are given conformal prediction sets their accuracy on tasks improves compared to fixed-size prediction sets with the same coverage guarantee. The results show that quantifying model uncertainty with conformal prediction is helpful for human-in-the-loop decision making and human-AI teams.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-cresswell24a, title = {Conformal Prediction Sets Improve Human Decision Making}, author = {Cresswell, Jesse C. and Sui, Yi and Kumar, Bhargava and Vouitsis, No\"{e}l}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {9439--9457}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/cresswell24a/cresswell24a.pdf}, url = {https://proceedings.mlr.press/v235/cresswell24a.html}, abstract = {In response to everyday queries, humans explicitly signal uncertainty and offer alternative answers when they are unsure. Machine learning models that output calibrated prediction sets through conformal prediction mimic this human behaviour; larger sets signal greater uncertainty while providing alternatives. In this work, we study the usefulness of conformal prediction sets as an aid for human decision making by conducting a pre-registered randomized controlled trial with conformal prediction sets provided to human subjects. With statistical significance, we find that when humans are given conformal prediction sets their accuracy on tasks improves compared to fixed-size prediction sets with the same coverage guarantee. The results show that quantifying model uncertainty with conformal prediction is helpful for human-in-the-loop decision making and human-AI teams.} }
Endnote
%0 Conference Paper %T Conformal Prediction Sets Improve Human Decision Making %A Jesse C. Cresswell %A Yi Sui %A Bhargava Kumar %A Noël Vouitsis %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-cresswell24a %I PMLR %P 9439--9457 %U https://proceedings.mlr.press/v235/cresswell24a.html %V 235 %X In response to everyday queries, humans explicitly signal uncertainty and offer alternative answers when they are unsure. Machine learning models that output calibrated prediction sets through conformal prediction mimic this human behaviour; larger sets signal greater uncertainty while providing alternatives. In this work, we study the usefulness of conformal prediction sets as an aid for human decision making by conducting a pre-registered randomized controlled trial with conformal prediction sets provided to human subjects. With statistical significance, we find that when humans are given conformal prediction sets their accuracy on tasks improves compared to fixed-size prediction sets with the same coverage guarantee. The results show that quantifying model uncertainty with conformal prediction is helpful for human-in-the-loop decision making and human-AI teams.
APA
Cresswell, J.C., Sui, Y., Kumar, B. & Vouitsis, N.. (2024). Conformal Prediction Sets Improve Human Decision Making. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:9439-9457 Available from https://proceedings.mlr.press/v235/cresswell24a.html.

Related Material