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].
Lower Complexity Adaptation for Empirical Entropic Optimal Transport
Authors: Michel Groppe, Shayan Hundrieser
JMLR 2024 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Additionally, we comment on computational aspects and complement our findings with Monte Carlo simulations. |
| Researcher Affiliation | Academia | Michel Groppe EMAIL Shayan Hundrieser EMAIL Institute for Mathematical Stochastics University of Göttingen Goldschmidtstraße 7, 37077 Göttingen, Germany |
| Pseudocode | No | The paper describes the Sinkhorn algorithm in text in Section 4, but does not present it as a structured pseudocode or algorithm block. |
| Open Source Code | Yes | The code used for our simulations can be found under https://gitlab.gwdg.de/michel.groppe/ eot-lca-simulations. |
| Open Datasets | No | The paper describes generating synthetic datasets, such as 'µ = Unif([0, 1]d1 {0}d1 d2 d1)' and 'ν = Unif([0, 1]d2 {0}d1 d2 d2)', but does not provide concrete access information like a URL, DOI, or repository for pre-generated data files. |
| Dataset Splits | No | The paper uses Monte Carlo simulations where data is generated from specified distributions for each repetition, rather than splitting a fixed dataset into training, testing, and validation sets. Therefore, it does not provide explicit dataset split information in the traditional sense. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used for running the simulations, such as GPU models, CPU types, or memory specifications. |
| Software Dependencies | No | The paper does not provide specific software dependencies or version numbers for libraries or solvers used in the simulations, beyond mentioning that code is available on GitLab. |
| Experiment Setup | Yes | for different sample sizes n by Monte Carlo simulations with 1000 repetitions. The empirical estimators b Tc,ε,n are approximated via the Sinkhorn algorithm such that the error of the first marginal is less than 10 8 w.r.t. the norm 1. The true value Tc,ε(µ, ν) is approximated using n = 6000 samples. ... we calculate the one-sample estimator Tc,ε(µ, ˆνn) and use n = 20000 samples to approximate Tc,ε(µ, ν). |