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 2402.07577cs.CL
收藏

主题建模作为多目标对比优化

Topic Modeling as Multi-Objective Contrastive Optimization

Institute of Data Science (IDS), National University of Singapore (NUS)(数据科学研究所(IDS),新加坡国立大学);Nanyang Technological University (NTU)(南洋理工大学);Carnegie Mellon University (CMU)(卡内基梅隆大学)

Thong Nguyen, Xiaobao Wu, Xinshuai Dong, Cong-Duy T Nguyen, See-Kiong Ng, Anh Tuan Luu

AI总结:

针对神经主题模型中文档级对比学习捕获低级信息及ELBO与对比损失冲突的问题,提出面向主题向量集合的对比学习方法,并将其转化为基于梯度的多目标优化问题以寻求帕累托平稳解,显著提升了主题连贯性、多样性与下游性能。

摘要:

近期的表示学习方法通过优化对数似然的证据下界(ELBO)与对比输入文档对的对比学习目标的加权线性组合,来增强神经主题模型。然而,文档级对比学习可能会捕获低级互信息(如词比例),从而干扰主题建模。此外,旨在记忆输入细节以提升重建质量的ELBO损失,与试图学习在输入文档间泛化的主题表示的对比损失之间存在潜在冲突。为解决这些问题,我们首先引入一种面向主题向量集合的新型对比学习方法,以捕获在一组输入文档间共享的有用语义。其次,我们明确地将对比主题建模转化为基于梯度的多目标优化问题,旨在实现平衡ELBO与对比目标之间权衡的帕累托平稳解。大量实验表明,我们的框架在主题连贯性、主题多样性和下游性能方面,持续产生性能更高的神经主题模型。

英文摘要:

Recent representation learning approaches enhance neural topic models by optimizing the weighted linear combination of the evidence lower bound (ELBO) of the log-likelihood and the contrastive learning objective that contrasts pairs of input documents. However, document-level contrastive learning might capture low-level mutual information, such as word ratio, which disturbs topic modeling. Moreover, there is a potential conflict between the ELBO loss that memorizes input details for better reconstruction quality, and the contrastive loss which attempts to learn topic representations that generalize among input documents. To address these issues, we first introduce a novel contrastive learning method oriented towards sets of topic vectors to capture useful semantics that are shared among a set of input documents. Secondly, we explicitly cast contrastive topic modeling as a gradient-based multi-objective optimization problem, with the goal of achieving a Pareto stationary solution that balances the trade-off between the ELBO and the contrastive objective. Extensive experiments demonstrate that our framework consistently produces higher-performing neural topic models in terms of topic coherence, topic diversity, and downstream performance.

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。

↑