arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.00680cs.LGcs.DSstat.ML
收藏

项目反应理论模型的可扩展学习

Scalable Learning of Item Response Theory Models

TU Dortmund(多特蒙德工业大学);Center for Data Science & Simulation (DoDaS)(数据科学与模拟中心)

Susanne Frick, Amer Krivošija, Alexander Munteanu

AI总结:

针对大规模数据下IRT模型计算效率与可扩展性不足的问题,研究利用IRT模型与逻辑回归的相似性,开发适配交替式IRT训练算法的核心集,实现IRT模型的可扩展学习。

摘要:

项目反应理论(Item Response Theory, IRT)模型旨在从表明作答质量的分类数据中,评估n名考生的潜在能力,以及m道测试题的潜在难度特征。经典心理测量评估基于相对少量的考生和题目,比如一个200名学生的班级解答包含10道题的考试。近期的全球大规模评估(如PISA)或互联网研究,可能导致参与者数量显著增加。此外,在机器学习领域中,算法扮演考生角色、数据分析问题扮演题目角色,n和m都可能变得非常大,对计算的效率和可扩展性构成挑战。为从大规模数据中学习IRT模型的潜在变量,我们利用这类模型与逻辑回归的相似性,后者可通过被称为coresets(核心集)的小型加权子集进行精确近似。我们开发了适用于交替式IRT训练算法的核心集,为大规模数据下的可扩展学习提供支持。

英文摘要:

Item Response Theory (IRT) models aim to assess latent abilities of $n$ examinees along with latent difficulty characteristics of $m$ test items from categorical data that indicates the quality of their corresponding answers. Classical psychometric assessments are based on a relatively small number of examinees and items, say a class of $200$ students solving an exam comprising $10$ problems. More recent global large scale assessments such as PISA, or internet studies, may lead to significantly increased numbers of participants. Additionally, in the context of Machine Learning where algorithms take the role of examinees and data analysis problems take the role of items, both $n$ and $m$ may become very large, challenging the efficiency and scalability of computations. To learn the latent variables in IRT models from large data, we leverage the similarity of these models to logistic regression, which can be approximated accurately using small weighted subsets called coresets. We develop coresets for their use in alternating IRT training algorithms, facilitating scalable learning from large data.

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。

↑