Preventing Overfitting via Sample Reweighting for Recommender System Incremental Update (Student Abstract)

Danni Peng, Xiaobo Hu, Anxiang Zeng, Jie Zhang

AAAI 2021 pp. 15863-15864

doi:10.1609/AAAI.V35I18.17928 /aaai/2021/peng2021aaai-preventing/

Abstract

Incremental update of recommender system models using only newly arrived data may easily cause the model to overfit to the current data. To address this issue without relying on historical data, we propose a sample reweighting method based on prediction performance of previous model on current data. The proposed method effectively alleviates the problem of overfitting and improves the performance of incremental update.

PDF AAAI Semantic Scholar

Cite

Text

Peng et al. "Preventing Overfitting via Sample Reweighting for Recommender System Incremental Update (Student Abstract)." AAAI Conference on Artificial Intelligence, 2021. doi:10.1609/AAAI.V35I18.17928

Markdown

[Peng et al. "Preventing Overfitting via Sample Reweighting for Recommender System Incremental Update (Student Abstract)." AAAI Conference on Artificial Intelligence, 2021.](https://mlanthology.org/aaai/2021/peng2021aaai-preventing/) doi:10.1609/AAAI.V35I18.17928

BibTeX

@inproceedings{peng2021aaai-preventing,
  title     = {{Preventing Overfitting via Sample Reweighting for Recommender System Incremental Update (Student Abstract)}},
  author    = {Peng, Danni and Hu, Xiaobo and Zeng, Anxiang and Zhang, Jie},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2021},
  pages     = {15863-15864},
  doi       = {10.1609/AAAI.V35I18.17928},
  url       = {https://mlanthology.org/aaai/2021/peng2021aaai-preventing/}
}