Universal Joint Approximation of Manifolds and Densities by Simple Injective Flows
Authors: Michael Puthawala, Matti Lassas, Ivan Dokmanic, Maarten De Hoop
ICML 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | In this work, we address approximation-theoretic properties of injective flows. We prove that under mild conditions these networks universally approximate probability measures supported on low-dimensional manifolds and describe how their design enables applications to inference and inverse problems. |
| Researcher Affiliation | Academia | 1Department of Computational and Applied Math, Rice University, Houston, TX, USA 2Department of Mathematics and Statistics, University of Helsinki, Finland 3Department of Mathematics and Computer Science, University of Basel, Basel, Switzerland. |
| Pseudocode | No | The paper does not contain any pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any statement about making its source code available or include links to code repositories. |
| Open Datasets | No | The paper is theoretical and does not conduct empirical studies using datasets, thus no information on training data availability or access is provided. |
| Dataset Splits | No | The paper is theoretical and does not conduct empirical studies, therefore no dataset split information (training, validation, test) is provided. |
| Hardware Specification | No | The paper is theoretical and does not describe experimental procedures, therefore no hardware specifications are mentioned. |
| Software Dependencies | No | The paper is theoretical and does not describe experimental procedures, therefore no software dependencies with version numbers are listed. |
| Experiment Setup | No | The paper is theoretical and does not describe experiments, therefore no experimental setup details such as hyperparameters or training settings are provided. |