學 術 演 講
學 術 演 講
題 目：Accelerating Item Factor Analysis on GPU and Vectorized Operations
時 間：民國110年12月6日 (星期一) 下午1：30
Item parameter estimation is a crucial step when conducting item factor analysis (IFA). Among existing estimation methods, marginal maximum likelihood (MML) seems to be the golden standard. However, fitting a high-dimensional IFA model by MML is still a challenging task. The current study demonstrates that with the help of GPU (graphics processing unit) and carefully designed vectorization, the computational time of MML could be largely reduced for large-scale IFA applications. In particular, a vectorized Metropolis-Hastings Robbins-Monro (VMHRM) algorithm is established with vectorized operations using contemporary deep learning libraries. Our numerical experiments show that VMHRM may run 33 times faster than its CPU version. When the number of factors is at least 5, VMHRM (on GPU) is much faster than the Bock-Aitkin expectation maximization, MHRM implemented by mirt (on CPU), and the importance-weighted autoencoder (on GPU). We believe that GPU computing will play a central role in large-scale psychometric modeling in the near future.