@inproceedings{b3a4f022cd784178977f1a8feef54591,
title = "HINT: Hierarchical Invertible Neural Transport for Density Estimation and Bayesian Inference",
abstract = " A large proportion of recent invertible neural architectures is based on a coupling block design. It operates by dividing incoming variables into two sub-spaces, one of which parameterizes an easily invertible (usually affine) transformation that is applied to the other. While the Jacobian of such a transformation is triangular, it is very sparse and thus may lack expressiveness. This work presents a simple remedy by noting that (affine) coupling can be repeated recursively within the resulting sub-spaces, leading to an efficiently invertible block with dense triangular Jacobian. By formulating our recursive coupling scheme via a hierarchical architecture, HINT allows sampling from a joint distribution p(y,x) and the corresponding posterior p(x|y) using a single invertible network. We demonstrate the power of our method for density estimation and Bayesian inference on a novel data set of 2D shapes in Fourier parameterization, which enables consistent visualization of samples for different dimensionalities. ",
keywords = "stat.ML, cs.AI, cs.LG",
author = "Jakob Kruse and Gianluca Detommaso and Robert Scheichl and Ullrich K{\"o}the",
year = "2021",
month = may,
day = "18",
doi = "10.1609/aaai.v35i9.16997",
language = "English",
volume = "35",
series = "Proceedings of the 35th AAAI Conference on Artificial Intelligence (AAAI-21)",
publisher = "Association for the Advancement of Artificial Intelligence (AAAI)",
pages = "8191--8199",
booktitle = "Proceedings of the 35th AAAI Conference on Artificial Intelligence (AAAI-21)",
edition = "9",
}