B1338
Title: Statistical inference using generative adversarial networks
Authors: Jingze Liu - Binghamton university (United States) [presenting]
Abstract: The potential of utilizing samples generated by generative adversarial networks (GANs) as a replacement for the conventional bootstrap resampling technique is investigated. Two procedures are introduced, one for low-dimensional and the other for high-dimensional cases, and their theoretical properties are demonstrated. Notably, the high-dimensional method has a convergence rate that is free of the curse of dimensionality. The preliminary simulation results are presented, which demonstrate that the GAN-based bootstrap method can produce reliable estimates of the variability and construct valid confidence intervals.