ICML 2021spotlight23 citations

Model Fusion for Personalized Learning

Thanh Chi Lam, Nghia Hoang, Bryan Kian Hsiang Low, Patrick Jaillet

Abstract

Production systems operating on a growing domain of analytic services often require generating warm-start solution models for emerging tasks with limited data. One potential approach to address this warm-start challenge is to adopt meta learning to generate a base model that can be adapted to solve unseen tasks with minimal fine-tuning. This however requires the training processes of previous solution models of existing tasks to be synchronized. This is not possible if these models were pre-trained separately on private data owned by different entities and cannot be synchronously re-trained. To accommodate for such scenarios, we develop a new personalized learning framework that synthesizes customized models for unseen tasks via fusion of independently pre-trained models of related tasks. We establish performance guarantee for the proposed framework and demonstrate its effectiveness on both synthetic and real datasets.

BibTeX
@InProceedings{pmlr-v139-lam21a,
  title = 	 {Model Fusion for Personalized Learning},
  author =       {Lam, Thanh Chi and Hoang, Nghia and Low, Bryan Kian Hsiang and Jaillet, Patrick},
  booktitle = 	 {Proceedings of the 38th International Conference on Machine Learning},
  pages = 	 {5948--5958},
  year = 	 {2021},
  editor = 	 {Meila, Marina and Zhang, Tong},
  volume = 	 {139},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {18--24 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v139/lam21a/lam21a.pdf},
  url = 	 {https://proceedings.mlr.press/v139/lam21a.html},
  abstract = 	 {Production systems operating on a growing domain of analytic services often require generating warm-start solution models for emerging tasks with limited data. One potential approach to address this warm-start challenge is to adopt meta learning to generate a base model that can be adapted to solve unseen tasks with minimal fine-tuning. This however requires the training processes of previous solution models of existing tasks to be synchronized. This is not possible if these models were pre-trained separately on private data owned by different entities and cannot be synchronously re-trained. To accommodate for such scenarios, we develop a new personalized learning framework that synthesizes customized models for unseen tasks via fusion of independently pre-trained models of related tasks. We establish performance guarantee for the proposed framework and demonstrate its effectiveness on both synthetic and real datasets.}
}
Model Fusion for Personalized Learning · ICML 2021