Spectral Mixture Kernels for Multi-Output Gaussian Processes

Authors: Gabriel Parra, Felipe Tobar

NeurIPS 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The proposed method is first validated on synthetic data and then compared to existing MOGP methods on two real-world examples. and We show two sets of experiments.
Researcher Affiliation Academia Gabriel Parra Department of Mathematical Engineering Universidad de Chile gparra@dim.uchile.cl Felipe Tobar Center for Mathematical Modeling Universidad de Chile ftobar@dim.uchile.cl
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code No The paper mentions implementing models in Tensorflow and GPflow, but does not provide concrete access to source code for the MOSM methodology described in this paper.
Open Datasets Yes The data can be obtained from www.cambermet.co.uk. and the sites therein. and The Jura dataset [3] contains, in addition to other geological data, the concentration of seven heavy metals in a region of 14.5 km2 of the Swiss Jura, and it is divided into a training set (259 locations) and a validation set (100 locations).
Dataset Splits Yes We chose N1 = 500 samples from the reference function in the interval [-20, 20], N2 = 400 samples from the derivative signal in the interval [-20, 0], and N3 = 400 samples from the delayed signal in the interval [-20, 0]. and We considered the normalised air temperature signal from... from where we randomly chose N = 1000 samples for training. and The Jura dataset [3] contains... divided into a training set (259 locations) and a validation set (100 locations).
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper mentions 'Tensorflow [18] using GPflow [19]' but does not provide specific version numbers for these software dependencies, which are required for reproducibility.
Experiment Setup Yes The NLL is then minimised with respect to Θ = {w(q) i , µ(q) i , Σ(q) i , θ(q) i , φ(q) i , σ2 i,noise}m,Q i=1,q=1, that is, the original parameters chosen to construct R(ω) in Section 3.1, plus the noise hyperparameters.