Acquisition Conditioned Oracle for Nongreedy Active Feature Acquisition

Michael Valancius, Maxwell Lennon, Junier Oliva
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:48957-48975, 2024.

Abstract

We develop novel methodology for active feature acquisition (AFA), the study of sequentially acquiring a dynamic subset of features that minimizes acquisition costs whilst still yielding accurate inference. The AFA framework can be useful in a myriad of domains, including health care applications where the cost of acquiring additional features for a patient (in terms of time, money, risk, etc.) can be weighed against the expected improvement to diagnostic performance. Previous approaches for AFA have employed either: deep learning RL techniques, which have difficulty training policies due to a complicated state and action space; deep learning surrogate generative models, which require modeling complicated multidimensional conditional distributions; or greedy policies, which cannot account for jointly informative feature acquisitions. We show that we can bypass many of these challenges with a novel, nonparametric oracle based approach, which we coin the acquisition conditioned oracle (ACO). Extensive experiments show the superiority of the ACO to state-of-the-art AFA methods when acquiring features for both predictions and general decision-making.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-valancius24a, title = {Acquisition Conditioned Oracle for Nongreedy Active Feature Acquisition}, author = {Valancius, Michael and Lennon, Maxwell and Oliva, Junier}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {48957--48975}, 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/valancius24a/valancius24a.pdf}, url = {https://proceedings.mlr.press/v235/valancius24a.html}, abstract = {We develop novel methodology for active feature acquisition (AFA), the study of sequentially acquiring a dynamic subset of features that minimizes acquisition costs whilst still yielding accurate inference. The AFA framework can be useful in a myriad of domains, including health care applications where the cost of acquiring additional features for a patient (in terms of time, money, risk, etc.) can be weighed against the expected improvement to diagnostic performance. Previous approaches for AFA have employed either: deep learning RL techniques, which have difficulty training policies due to a complicated state and action space; deep learning surrogate generative models, which require modeling complicated multidimensional conditional distributions; or greedy policies, which cannot account for jointly informative feature acquisitions. We show that we can bypass many of these challenges with a novel, nonparametric oracle based approach, which we coin the acquisition conditioned oracle (ACO). Extensive experiments show the superiority of the ACO to state-of-the-art AFA methods when acquiring features for both predictions and general decision-making.} }
Endnote
%0 Conference Paper %T Acquisition Conditioned Oracle for Nongreedy Active Feature Acquisition %A Michael Valancius %A Maxwell Lennon %A Junier Oliva %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-valancius24a %I PMLR %P 48957--48975 %U https://proceedings.mlr.press/v235/valancius24a.html %V 235 %X We develop novel methodology for active feature acquisition (AFA), the study of sequentially acquiring a dynamic subset of features that minimizes acquisition costs whilst still yielding accurate inference. The AFA framework can be useful in a myriad of domains, including health care applications where the cost of acquiring additional features for a patient (in terms of time, money, risk, etc.) can be weighed against the expected improvement to diagnostic performance. Previous approaches for AFA have employed either: deep learning RL techniques, which have difficulty training policies due to a complicated state and action space; deep learning surrogate generative models, which require modeling complicated multidimensional conditional distributions; or greedy policies, which cannot account for jointly informative feature acquisitions. We show that we can bypass many of these challenges with a novel, nonparametric oracle based approach, which we coin the acquisition conditioned oracle (ACO). Extensive experiments show the superiority of the ACO to state-of-the-art AFA methods when acquiring features for both predictions and general decision-making.
APA
Valancius, M., Lennon, M. & Oliva, J.. (2024). Acquisition Conditioned Oracle for Nongreedy Active Feature Acquisition. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:48957-48975 Available from https://proceedings.mlr.press/v235/valancius24a.html.

Related Material