Stabilizing Invertible Neural Networks Using Mixture Models

DOI10.1088/1361-6420/abe928OpenAlexW3129907242MaRDI QIDQ5002576

Paul Hagemann, Sebastian Neumayer

Publication date: 28 July 2021

Published in: Inverse Problems (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2009.02994

zbMATH Keywords

inverse problemsnumerical stabilitymixture modelsinvertible neural networksmultimodal problems

Mathematics Subject Classification ID

Artificial intelligence (68Txx) Numerical analysis (65-XX) Probability theory on algebraic and topological structures (60Bxx)

Related Items (6)

Invertible neural networks versus MCMC for posterior reconstruction in grazing incidence X-ray fluorescence ⋮ Stochastic Normalizing Flows for Inverse Problems: A Markov Chains Viewpoint ⋮ Stable parameterization of continuous and piecewise-linear functions ⋮ An introduction to deep generative modeling ⋮ Approximation of Lipschitz Functions Using Deep Spline Neural Networks ⋮ Convolutional proximal neural networks and plug-and-play algorithms

Uses Software

• Adam
• Glow
• torchdiffeq
• Wasserstein GAN
• DeLiGAN

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Parseval proximal neural networks
• General state space Markov chains and MCMC algorithms
• Probability theory. Translated from the German by Robert B. Burckel
• Optimal transport for applied mathematicians. Calculus of variations, PDEs, and modeling
• Hilbert space embeddings and metrics on probability measures
• On Choosing and Bounding Probability Metrics
• An Introduction to Variational Autoencoders
• Wahrscheinlichkeitstheorie

This page was built for publication: Stabilizing Invertible Neural Networks Using Mixture Models