Testing and Learning on Distributions with Symmetric Noise Invariance

Authors: Ho Chung Law, Christopher Yau, Dino Sejdinovic

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

Reproducibility Variable Result LLM Response
Research Type Experimental Section 5 provides experiments on synthetic and real data, before concluding in section 6.
Researcher Affiliation Academia Ho Chung Leon Law Department of Statistics University Of Oxford hlaw@stats.ox.ac.uk Christopher Yau Centre for Computational Biology University of Birmingham c.yau@bham.ac.uk Dino Sejdinovic Department of Statistics University Of Oxford dino.sejdinovic@stats.ox.ac.uk
Pseudocode Yes To do this, we can construct a neural network (NN) with special activation functions, pooling layers as shown in Algorithm D.1 and Figure D.1 in the Appendix.
Open Source Code No The paper does not provide any explicit statements or links about the availability of source code for the methodology described.
Open Datasets Yes To demonstrate the phase features invariance to SPD noise component, we use the Aerosol MISR dataset also studied by [24] and [25] and consider a situation with covariate shift [18] on distribution inputs: the testing data is impaired by additive SPD components different to that in the training data.
Dataset Splits Yes For experiments, we use approximately 9000 bags for training, and 3000 bags each for validation and testing, keeping those of multiple lines of sight in the same set.
Hardware Specification No The paper does not provide specific hardware details (e.g., exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiments.
Experiment Setup Yes In the SME test, we take the number of test locations J to be 10, and use 20% of the samples to optimise the test locations.