@inproceedings{gao-etal-2020-setconv,
    title = "{S}et{C}onv: {A} {N}ew {A}pproach for {L}earning from {I}mbalanced {D}ata",
    author = "Gao, Yang  and
      Li, Yi-Fan  and
      Lin, Yu  and
      Aggarwal, Charu  and
      Khan, Latifur",
    editor = "Webber, Bonnie  and
      Cohn, Trevor  and
      He, Yulan  and
      Liu, Yang",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2020.emnlp-main.98/",
    doi = "10.18653/v1/2020.emnlp-main.98",
    pages = "1284--1294",
    abstract = "For many real-world classification problems, e.g., sentiment classification, most existing machine learning methods are biased towards the majority class when the Imbalance Ratio (IR) is high. To address this problem, we propose a set convolution (SetConv) operation and an episodic training strategy to extract a single representative for each class, so that classifiers can later be trained on a balanced class distribution. We prove that our proposed algorithm is permutation-invariant despite the order of inputs, and experiments on multiple large-scale benchmark text datasets show the superiority of our proposed framework when compared to other SOTA methods."
}