Model Selection for Behavioral Learning Data and Applications to Contextual Bandits
Abstract
Learning for animals or humans is the process that leads to behaviors better adapted to the environment. This process highly depends on the individual that learns and is usually observed only through the individual’s actions. This article presents ways to use this individual behavioral data to find the model that best explains how the individual learns. We propose two model selection methods: a general hold-out procedure and an AIC-type criterion, both adapted to non-stationary dependent data. We provide theoretical error bounds for these methods that are close to those of the standard i.i.d. case. To compare these approaches, we apply them to contextual bandit models and illustrate their use on both synthetic and experimental learning data in a human categorization task.
Cite
Text
Aubert et al. "Model Selection for Behavioral Learning Data and Applications to Contextual Bandits." Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, 2025.Markdown
[Aubert et al. "Model Selection for Behavioral Learning Data and Applications to Contextual Bandits." Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, 2025.](https://mlanthology.org/aistats/2025/aubert2025aistats-model/)BibTeX
@inproceedings{aubert2025aistats-model,
title = {{Model Selection for Behavioral Learning Data and Applications to Contextual Bandits}},
author = {Aubert, Julien and Köhler, Louis and Lehéricy, Luc and Mezzadri, Giulia and Reynaud-Bouret, Patricia},
booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics},
year = {2025},
pages = {1126-1134},
volume = {258},
url = {https://mlanthology.org/aistats/2025/aubert2025aistats-model/}
}