A Locally Adaptive Normal Distribution
Authors: Georgios Arvanitidis, Lars K. Hansen, Søren Hauberg
NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate the efficiency of the LAND to fit non-trivial probability distributions over both synthetic data, and EEG measurements of human sleep. and The usefulness of the model is verified on both synthetic data and EEG measurements of human sleep stages. and In this section we present both synthetic and real experiments to demonstrate the advantages of the LAND. |
| Researcher Affiliation | Academia | Georgios Arvanitidis, Lars Kai Hansen and Søren Hauberg Technical University of Denmark, Lyngby, Denmark DTU Compute, Section for Cognitive Systems {gear,lkai,sohau}@dtu.dk |
| Pseudocode | Yes | Algorithm 1 LAND maximum likelihood |
| Open Source Code | No | The paper does not contain any explicit statements about releasing source code for the described methodology nor does it provide any links to a code repository. |
| Open Datasets | Yes | We consider electro-encephalography (EEG) measurements of human sleep from 10 subjects, part of the Physio Net database [11, 7, 5]. |
| Dataset Splits | No | The paper describes generating synthetic data and using EEG measurements, but it does not specify explicit train/validation/test dataset splits (e.g., percentages, sample counts, or predefined splits) that would allow for reproduction of the data partitioning methodology. |
| Hardware Specification | No | The paper does not provide any specific details regarding the hardware used for running the experiments, such as GPU/CPU models, memory, or other computational resources. |
| Software Dependencies | No | The paper mentions using 'EEGLAB' as an open-source toolbox but does not provide specific version numbers for it or any other software dependencies like programming languages or libraries. |
| Experiment Setup | Yes | In all experiments we use S = 3000 samples in the Monte Carlo integration. and Furthermore, we choose σ as small as possible, while ensuring that the manifold is smooth enough that geodesics can be computed numerically. and From the results in Table 1 we observe that for σ = 1 for all the subjects. and Non-Negative Matrix Factorization (10 random starts) and Algorithm 1 LAND maximum likelihood which includes stepsizes αµ, αA. |