DeepAI AI Chat
Log In Sign Up

Benchmarking Invertible Architectures on Inverse Problems

by   Jakob Kruse, et al.

Recent work demonstrated that flow-based invertible neural networks are promising tools for solving ambiguous inverse problems. Following up on this, we investigate how ten invertible architectures and related models fare on two intuitive, low-dimensional benchmark problems, obtaining the best results with coupling layers and simple autoencoders. We hope that our initial efforts inspire other researchers to evaluate their invertible architectures in the same setting and put forth additional benchmarks, so our evaluation may eventually grow into an official community challenge.


page 1

page 2

page 3

page 4


Convergence and Recovery Guarantees of Unsupervised Neural Networks for Inverse Problems

Neural networks have become a prominent approach to solve inverse proble...

An artificial neural network approximation for Cauchy inverse problems

A novel artificial neural network method is proposed for solving Cauchy ...

Provably Convergent Algorithms for Solving Inverse Problems Using Generative Models

The traditional approach of hand-crafting priors (such as sparsity) for ...

Deep Learning Methods for Solving Linear Inverse Problems: Research Directions and Paradigms

The linear inverse problem is fundamental to the development of various ...

On the Robustness of Normalizing Flows for Inverse Problems in Imaging

Conditional normalizing flows can generate diverse image samples for sol...

Two Birds with One Stone: Investigating Invertible Neural Networks for Inverse Problems in Morphology

Most problems in natural language processing can be approximated as inve...