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].
Variational excess risk bound for general state space models
Authors: Elisabeth Gassiat, Sylvain Le Corff
TMLR 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | Our theoretical results provide the first excess risk bounds in a context of VAE for state space models. The proposed upper bound has the same behaviour in n as the ones obtained in the classical statistical literature for parametric models, in which 1/n is the parametric rate and log n factors come from the concentration of empirical measures. |
| Researcher Affiliation | Academia | Elisabeth Gassiat EMAIL Université Paris-Saclay, CNRS Laboratoire de mathématiques d Orsay 91405, Orsay, France Sylvain Le Corff EMAIL LPSM Sorbonne Université, UMR CNRS 8001 Paris, France |
| Pseudocode | No | The paper describes theoretical results and mathematical proofs. While it mentions optimization methods like ADAM in the context of practical implementation, it does not provide any pseudocode or algorithm blocks for its own methodology or findings. |
| Open Source Code | No | The paper does not contain any explicit statements about releasing source code, nor does it provide links to any code repositories or mention code in supplementary materials. |
| Open Datasets | No | The paper is theoretical and focuses on general state space models and properties of data distributions. It does not mention the use of any specific publicly available or open datasets for empirical evaluation. |
| Dataset Splits | No | As the paper focuses on theoretical analysis and does not conduct experiments on specific datasets, there is no information provided regarding training/test/validation dataset splits. |
| Hardware Specification | No | The paper is theoretical in nature, providing mathematical results and bounds. It does not describe any experimental setup or specify the hardware used for any computational work. |
| Software Dependencies | No | The paper is primarily theoretical and does not detail specific software dependencies or version numbers for its own development or implementation. It references 'ADAM Kingma & Ba (2015)' as a common approach for approximate estimators but does not list it as a software dependency for its own work. |
| Experiment Setup | No | The paper presents theoretical results and mathematical proofs, focusing on oracle inequalities and bounds. It does not include an experimental section with specific hyperparameters, training configurations, or system-level settings. |