Inference for determinantal point processes without spectral knowledge
Authors: Rémi Bardenet, Michalis Titsias RC AUEB
NeurIPS 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In Section 5, we experimentally validate our results, before discussing their breadth in Section 6. 5 Experiments |
| Researcher Affiliation | Academia | R emi Bardenet CNRS & CRISt AL UMR 9189, Univ. Lille, France remi.bardenet@gmail.com Michalis K. Titsias Department of Informatics Athens Univ. of Economics and Business, Greece mtitsias@aueb.gr |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any concrete access information (e.g., repository link, explicit statement of code release) for the source code of the described methodology. |
| Open Datasets | Yes | Here, we consider a real dataset of spatial patterns of nerve bers in diabetic patients. These bers become more clustered as diabetes progresses [22]. The dataset consists of 7 samples collected from diabetic patients at di erent stages of diabetic neuropathy and one healthy subject. |
| Dataset Splits | Yes | We follow the experimental setup used in [7] and we split the total samples into two classes: Normal/Mildly Diabetic and Moderately/Severely Diabetic. The brst class contains three samples and the second one the remaining four. |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper mentions 'Our MATLAB implementation' but does not specify its version or any other software dependencies with version numbers. |
| Experiment Setup | Yes | We sample a synthetic dataset using (κ, α, ϵ) = (1000, 0.5, 1)... To limit the range of κ, we choose for (log κ, log α, log ϵ) a wide uniform prior over [200, 2000] [ 10, 10] [ 10, 10]. We start each iteration with m = 20 pseudo-inputs, and increase it by 10 and re-optimize when the acceptance decision cannot be made. Removing a burn-in sample of size 1000... and ...we train the DPPs under di erent approximations having m {50, 100, 200, 400, 800, 1200} inducing variables... |