Asymptotic Bayes Risk of Semi-Supervised Multitask Learning on Gaussian Mixture
Abstract
The article considers semi-supervised multitask learning on a Gaussian mixture model (GMM). Using methods from statistical physics, we compute the asymptotic Bayes risk of each task in the regime of large datasets in high dimension, from which we analyze the role of task similarity in learning and evaluate the performance gain when tasks are learned together rather than separately. In the supervised case, we derive a simple algorithm that attains the Bayes optimal performance.
Cite
Text
Nguyen and Couillet. "Asymptotic Bayes Risk of Semi-Supervised Multitask Learning on Gaussian Mixture." Artificial Intelligence and Statistics, 2023.Markdown
[Nguyen and Couillet. "Asymptotic Bayes Risk of Semi-Supervised Multitask Learning on Gaussian Mixture." Artificial Intelligence and Statistics, 2023.](https://mlanthology.org/aistats/2023/nguyen2023aistats-asymptotic/)BibTeX
@inproceedings{nguyen2023aistats-asymptotic,
title = {{Asymptotic Bayes Risk of Semi-Supervised Multitask Learning on Gaussian Mixture}},
author = {Nguyen, Minh-Toan and Couillet, Romain},
booktitle = {Artificial Intelligence and Statistics},
year = {2023},
pages = {5063-5078},
volume = {206},
url = {https://mlanthology.org/aistats/2023/nguyen2023aistats-asymptotic/}
}