i-flow: High-dimensional Integration and Sampling with Normalizing Flows

01/15/2020
by   Christina Gao, et al.
0

In many fields of science, high-dimensional integration is required. Numerical methods have been developed to evaluate these complex integrals. We introduce the code i-flow, a python package that performs high-dimensional numerical integration utilizing normalizing flows. Normalizing flows are machine-learned, bijective mappings between two distributions. i-flow can also be used to sample random points according to complicated distributions in high dimensions. We compare i-flow to other algorithms for high-dimensional numerical integration and show that i-flow outperforms them for high dimensional correlated integrals.

READ FULL TEXT
research
09/02/2019

Randomized methods to characterize large-scale vortical flow network

We demonstrate the effective use of randomized methods for linear algebr...
research
12/14/2021

The high-dimensional asymptotics of first order methods with random data

We study a class of deterministic flows in ℝ^d× k, parametrized by a ran...
research
05/15/2023

Non-Separable Multi-Dimensional Network Flows for Visual Computing

Flows in networks (or graphs) play a significant role in numerous comput...
research
07/08/2023

Fast Empirical Scenarios

We seek to extract a small number of representative scenarios from large...
research
03/23/2022

Adjoint characteristics for Eulerian two-dimensional supersonic flow

The formal expressions, in terms of local flow variables, defining the a...
research
09/18/2023

The NFLikelihood: an unsupervised DNNLikelihood from Normalizing Flows

We propose the NFLikelihood, an unsupervised version, based on Normalizi...
research
03/02/2022

Flow-based density of states for complex actions

Emerging sampling algorithms based on normalizing flows have the potenti...

Please sign up or login with your details

Forgot password? Click here to reset