https://scholars.lib.ntu.edu.tw/handle/123456789/580910
標題: | Order-free learning alleviating exposure bias in multi-label classification | 作者: | Tsai C.-P HUNG-YI LEE |
關鍵字: | Decoding; Large dataset; Recurrent neural networks; Benchmark datasets; Generalization capability; Label combinations; Label dependencies; Multi label classification; Multiple labels; Recurrent neural network (RNN); Sequence prediction; Classification (of information) | 公開日期: | 2020 | 起(迄)頁: | 6038-6045 | 來源出版物: | AAAI 2020 - 34th AAAI Conference on Artificial Intelligence | 摘要: | Multi-label classification (MLC) assigns multiple labels to each sample. Prior studies show that MLC can be transformed to a sequence prediction problem with a recurrent neural network (RNN) decoder to model the label dependency. However, training a RNN decoder requires a predefined order of labels, which is not directly available in the MLC specification. Besides, RNN thus trained tends to overfit the label combinations in the training set and have difficulty generating unseen label sequences. In this paper, we propose a new framework for MLC which does not rely on a predefined label order and thus alleviates exposure bias. The experimental results on three multi-label classification benchmark datasets show that our method outperforms competitive baselines by a large margin. We also find the proposed approach has a higher probability of generating label combinations not seen during training than the baseline models. The result shows that the proposed approach has better generalization capability. Copyright ? 2020, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. |
URI: | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85090400953&partnerID=40&md5=9feaa2fc167163ceb7ccf95fde05d37c https://scholars.lib.ntu.edu.tw/handle/123456789/580910 |
顯示於: | 電機工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。