arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2304.11045cs.LGcs.AIcs.IR
收藏

轻量级深度极限多标签分类

Light-weight Deep Extreme Multilabel Classification

IIIT Hyderabad(海得拉巴国际信息技术学院);Microsoft India(微软印度公司)

Istasis Mishra, Arpan Dasgupta, Pratik Jawanpuria, Bamdev Mishra, Pawan Kumar

AI总结:

本文提出轻量深度 XML 方法 LightDXML,通过标签嵌入负采样与三阶段循环训练移除重排序器,在效率媲美树方法的同时达到深度模型精度并改善尾标签预测。

摘要:

极限多标签(XML)分类是指涉及大量标签的监督式多标签学习任务。因此,分类器随标签维度增加而具备的可扩展性是一个重要考量。本文提出一种名为 LightDXML 的方法,它改进了近期开发的基于深度学习的 XML 框架:使用标签嵌入而非特征嵌入进行负采样,并循环迭代三个主要阶段:(1)标签嵌入的代理训练;(2)为负采样筛选标签;(3)利用负样本训练最终分类器。因此,LightDXML 还去除了对重排序模块的需求,从而进一步节省时间和内存开销。所提方法实现了两全其美:其训练时间、模型规模和预测时间与基于树的方法相当或更优,同时获得了与基于深度学习的方法相当的、明显更好的预测准确率。此外,在部分大型数据集上,与大多数最先进的 XML 方法相比,该方法在尾标签预测准确率上取得最佳结果。(论文发表于 IJCNN 2023;部分资金来自印度海得拉巴 IIIT 的 MAPG 基金和 IIIT Seed 基金;代码位于 https://github.com/misterpawan/LightDXML)

英文摘要:

Extreme multi-label (XML) classification refers to the task of supervised multi-label learning that involves a large number of labels. Hence, scalability of the classifier with increasing label dimension is an important consideration. In this paper, we develop a method called LightDXML which modifies the recently developed deep learning based XML framework by using label embeddings instead of feature embedding for negative sampling and iterating cyclically through three major phases: (1) proxy training of label embeddings (2) shortlisting of labels for negative sampling and (3) final classifier training using the negative samples. Consequently, LightDXML also removes the requirement of a re-ranker module, thereby, leading to further savings on time and memory requirements. The proposed method achieves the best of both worlds: while the training time, model size and prediction times are on par or better compared to the tree-based methods, it attains much better prediction accuracy that is on par with the deep learning based methods. Moreover, the proposed approach achieves the best tail-label prediction accuracy over most state-of-the-art XML methods on some of the large datasets\footnote{accepted in IJCNN 2023, partial funding from MAPG grant and IIIT Seed grant at IIIT, Hyderabad, India. Code: \url{https://github.com/misterpawan/LightDXML}

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑