Neural Active Learning Meets the Partial Monitoring Framework

Maxime Heuillet, Ola Ahmad, Audrey Durand
Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, PMLR 244:1621-1639, 2024.

Abstract

We focus on the online-based active learning (OAL) setting where an agent operates over a stream of observations and trades-off between the costly acquisition of information (labelled observations) and the cost of prediction errors. We propose a novel foundation for OAL tasks based on partial monitoring, a theoretical framework specialized in online learning from partially informative actions. We show that previously studied binary and multi-class OAL tasks are instances of partial monitoring. We expand the real-world potential of OAL by introducing a new class of cost-sensitive OAL tasks. We propose NeuralCBP, the first PM strategy that accounts for predictive uncertainty with deep neural networks. Our extensive empirical evaluation on open source datasets shows that NeuralCBP has competitive performance against state-of-the-art baselines on multiple binary, multi-class and cost-sensitive OAL tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v244-heuillet24a, title = {Neural Active Learning Meets the Partial Monitoring Framework}, author = {Heuillet, Maxime and Ahmad, Ola and Durand, Audrey}, booktitle = {Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence}, pages = {1621--1639}, year = {2024}, editor = {Kiyavash, Negar and Mooij, Joris M.}, volume = {244}, series = {Proceedings of Machine Learning Research}, month = {15--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v244/main/assets/heuillet24a/heuillet24a.pdf}, url = {https://proceedings.mlr.press/v244/heuillet24a.html}, abstract = {We focus on the online-based active learning (OAL) setting where an agent operates over a stream of observations and trades-off between the costly acquisition of information (labelled observations) and the cost of prediction errors. We propose a novel foundation for OAL tasks based on partial monitoring, a theoretical framework specialized in online learning from partially informative actions. We show that previously studied binary and multi-class OAL tasks are instances of partial monitoring. We expand the real-world potential of OAL by introducing a new class of cost-sensitive OAL tasks. We propose NeuralCBP, the first PM strategy that accounts for predictive uncertainty with deep neural networks. Our extensive empirical evaluation on open source datasets shows that NeuralCBP has competitive performance against state-of-the-art baselines on multiple binary, multi-class and cost-sensitive OAL tasks.} }
Endnote
%0 Conference Paper %T Neural Active Learning Meets the Partial Monitoring Framework %A Maxime Heuillet %A Ola Ahmad %A Audrey Durand %B Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2024 %E Negar Kiyavash %E Joris M. Mooij %F pmlr-v244-heuillet24a %I PMLR %P 1621--1639 %U https://proceedings.mlr.press/v244/heuillet24a.html %V 244 %X We focus on the online-based active learning (OAL) setting where an agent operates over a stream of observations and trades-off between the costly acquisition of information (labelled observations) and the cost of prediction errors. We propose a novel foundation for OAL tasks based on partial monitoring, a theoretical framework specialized in online learning from partially informative actions. We show that previously studied binary and multi-class OAL tasks are instances of partial monitoring. We expand the real-world potential of OAL by introducing a new class of cost-sensitive OAL tasks. We propose NeuralCBP, the first PM strategy that accounts for predictive uncertainty with deep neural networks. Our extensive empirical evaluation on open source datasets shows that NeuralCBP has competitive performance against state-of-the-art baselines on multiple binary, multi-class and cost-sensitive OAL tasks.
APA
Heuillet, M., Ahmad, O. & Durand, A.. (2024). Neural Active Learning Meets the Partial Monitoring Framework. Proceedings of the Fortieth Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 244:1621-1639 Available from https://proceedings.mlr.press/v244/heuillet24a.html.

Related Material