Model selection for behavioral learning data and applications to contextual bandits

Julien Aubert, Louis Köhler, Luc Lehéricy, Giulia Mezzadri, Patricia Reynaud-Bouret
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:1126-1134, 2025.

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 this Paper

BibTeX
@InProceedings{pmlr-v258-aubert25a, title = {Model selection for behavioral learning data and applications to contextual bandits}, author = {Aubert, Julien and K{\"o}hler, Louis and Leh{\'e}ricy, Luc and Mezzadri, Giulia and Reynaud-Bouret, Patricia}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {1126--1134}, year = {2025}, editor = {Li, Yingzhen and Mandt, Stephan and Agrawal, Shipra and Khan, Emtiyaz}, volume = {258}, series = {Proceedings of Machine Learning Research}, month = {03--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v258/main/assets/aubert25a/aubert25a.pdf}, url = {https://proceedings.mlr.press/v258/aubert25a.html}, 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.} }
Endnote
%0 Conference Paper %T Model selection for behavioral learning data and applications to contextual bandits %A Julien Aubert %A Louis Köhler %A Luc Lehéricy %A Giulia Mezzadri %A Patricia Reynaud-Bouret %B Proceedings of The 28th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2025 %E Yingzhen Li %E Stephan Mandt %E Shipra Agrawal %E Emtiyaz Khan %F pmlr-v258-aubert25a %I PMLR %P 1126--1134 %U https://proceedings.mlr.press/v258/aubert25a.html %V 258 %X 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.
APA
Aubert, J., Köhler, L., Lehéricy, L., Mezzadri, G. & Reynaud-Bouret, P.. (2025). Model selection for behavioral learning data and applications to contextual bandits. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:1126-1134 Available from https://proceedings.mlr.press/v258/aubert25a.html.

Related Material