Fitting item response theory models using deep learning computational frameworks
Nanyu Luo, Feng Ji, Yuting Han, Jinbo He, Xiaoya Zhang · 2024
PyTorch and TensorFlow are two widely adopted, modern deep learning frameworksthat offer comprehensive computation libraries for deep learning models. We illustratehow to utilize these deep learning computational platforms and infrastructure to estimate aclass of popular psychometric models, dichotomous and polytomous Item Response Theory(IRT) models, along with their multidimensional extensions. Through simulation studies,the estimation performance on the simulated datasets demonstrates low mean square errorand bias for model parameters. We discuss the potential of integrating modern deeplearning tools and views into psychometric research.