Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Rates of convergence for density estimation with generative adversarial networks
Authors: Nikita Puchkin, Sergey Samsonov, Denis Belomestny, Eric Moulines, Alexey Naumov
JMLR 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | In this work we undertake a thorough study of the non-asymptotic properties of the vanilla generative adversarial networks (GANs). We prove an oracle inequality for the Jensen Shannon (JS) divergence between the underlying density p and the GAN estimate with a significantly better statistical error term compared to the previously known results. ... We prove (Theorem 1) a sharp oracle inequality... We show that the result of Theorem 2 is minimax optimal up to a logarithmic factor. Namely, we prove that for any estimate bp, it holds that... |
| Researcher Affiliation | Academia | Nikita Puchkin EMAIL HSE University, Russian Federation Institute for Information transmission Problems, Russian Federation Sergey Samsonov EMAIL HSE University, Russian Federation Institute for Information transmission Problems, Russian Federation Denis Belomestny EMAIL Duisburg-Essen University, Germany HSE University, Russian Federation Eric Moulines EMAIL Ecole Polytechnique, France Mohamed Bin Zayed University of AI, United Arab Emirates Alexey Naumov EMAIL HSE University, Russian Federation Steklov Mathematical Institute of Russian Academy of Sciences, Russian Federation |
| Pseudocode | No | No explicit pseudocode or algorithm blocks are present in the paper. The methodology is described mathematically through theorems and proofs. |
| Open Source Code | No | The paper focuses on theoretical analysis and does not mention the release of any open-source code for the described methodology. |
| Open Datasets | No | The paper uses X = Y = [0, 1]d as a theoretical domain for its nonparametric density estimation problem and does not mention the use or availability of any specific public datasets for experimental evaluation. |
| Dataset Splits | No | The paper is theoretical and does not conduct experiments on specific datasets, therefore, no information regarding dataset splits is provided. |
| Hardware Specification | No | The paper presents theoretical results and does not describe any experimental hardware specifications. |
| Software Dependencies | No | The paper is theoretical and focuses on mathematical derivations; it does not list any software dependencies with specific version numbers. |
| Experiment Setup | No | The paper provides a theoretical analysis of generative adversarial networks and does not include details about an experimental setup, hyperparameters, or training configurations. |