Weighted Contrastive Learning With False Negative Control to Help Long-tailed Product Classification
Tianqi Wang, Lei Chen, Xiaodan Zhu, Younghun Lee, Jing Gao
Abstract
Item categorization (IC) aims to classify product descriptions into leaf nodes in a categorical taxonomy, which is a key technology used in a wide range of applications. Along with the fact that most datasets often has a long-tailed distribution, classification performances on tail labels tend to be poor due to scarce supervision, causing many issues in real-life applications. To address IC task’s long-tail issue, K-positive contrastive loss (KCL) is proposed on image classification task and can be applied on the IC task when using text-based contrastive learning, e.g., SimCSE. However, one shortcoming of using KCL has been neglected in previous research: false negative (FN) instances may harm the KCL’s representation learning. To address the FN issue in the KCL, we proposed to re-weight the positive pairs in the KCL loss with a regularization that the sum of weights should be constrained to K+1 as close as possible. After controlling FN instances with the proposed method, IC performance has been further improved and is superior to other LT-addressing methods.
BibTeX
@inproceedings{wang-etal-2023-weighted,
title = "Weighted Contrastive Learning With False Negative Control to Help Long-tailed Product Classification",
author = "Wang, Tianqi and
Chen, Lei and
Zhu, Xiaodan and
Lee, Younghun and
Gao, Jing",
editor = "Sitaram, Sunayana and
Beigman Klebanov, Beata and
Williams, Jason D",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 5: Industry Track)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-industry.55/",
doi = "10.18653/v1/2023.acl-industry.55",
pages = "574--580"
}