Mode Estimation with Partial Feedback

Charles Arnal, Vivien Cabannes, Vianney Perchet
Proceedings of Thirty Seventh Conference on Learning Theory, PMLR 247:219-220, 2024.

Abstract

The combination of lightly supervised pre-training and online fine-tuning has played a key role in recent AI developments. These new learning pipelines call for new theoretical frameworks. In this paper, we formalize key aspects of weakly supervised and active learning with a simple problem: the estimation of the mode of a distribution with partial feedback. We showcase how entropy coding allows for optimal information acquisition from partial feedback, develop coarse sufficient statistics for mode identification, and adapt bandit algorithms to our new setting. Finally, we combine those contributions into a statistically and computationally efficient solution to our original problem.

Cite this Paper

BibTeX
@InProceedings{pmlr-v247-arnal24a, title = {Mode Estimation with Partial Feedback}, author = {Arnal, Charles and Cabannes, Vivien and Perchet, Vianney}, booktitle = {Proceedings of Thirty Seventh Conference on Learning Theory}, pages = {219--220}, year = {2024}, editor = {Agrawal, Shipra and Roth, Aaron}, volume = {247}, series = {Proceedings of Machine Learning Research}, month = {30 Jun--03 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v247/arnal24a/arnal24a.pdf}, url = {https://proceedings.mlr.press/v247/arnal24a.html}, abstract = { The combination of lightly supervised pre-training and online fine-tuning has played a key role in recent AI developments. These new learning pipelines call for new theoretical frameworks. In this paper, we formalize key aspects of weakly supervised and active learning with a simple problem: the estimation of the mode of a distribution with partial feedback. We showcase how entropy coding allows for optimal information acquisition from partial feedback, develop coarse sufficient statistics for mode identification, and adapt bandit algorithms to our new setting. Finally, we combine those contributions into a statistically and computationally efficient solution to our original problem. } }
Endnote
%0 Conference Paper %T Mode Estimation with Partial Feedback %A Charles Arnal %A Vivien Cabannes %A Vianney Perchet %B Proceedings of Thirty Seventh Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2024 %E Shipra Agrawal %E Aaron Roth %F pmlr-v247-arnal24a %I PMLR %P 219--220 %U https://proceedings.mlr.press/v247/arnal24a.html %V 247 %X The combination of lightly supervised pre-training and online fine-tuning has played a key role in recent AI developments. These new learning pipelines call for new theoretical frameworks. In this paper, we formalize key aspects of weakly supervised and active learning with a simple problem: the estimation of the mode of a distribution with partial feedback. We showcase how entropy coding allows for optimal information acquisition from partial feedback, develop coarse sufficient statistics for mode identification, and adapt bandit algorithms to our new setting. Finally, we combine those contributions into a statistically and computationally efficient solution to our original problem.
APA
Arnal, C., Cabannes, V. & Perchet, V.. (2024). Mode Estimation with Partial Feedback. Proceedings of Thirty Seventh Conference on Learning Theory, in Proceedings of Machine Learning Research 247:219-220 Available from https://proceedings.mlr.press/v247/arnal24a.html.

Related Material