Theoretical guarantees for EM under misspecified Gaussian mixture models

Authors: Raaz Dwivedi, nhật Hồ, Koulik Khamaru, Martin J. Wainwright, Michael I. Jordan

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We validate our theoretical findings in different cases via several numerical examples.
Researcher Affiliation Collaboration Raaz Dwivedi Nhat Ho Koulik Khamaru UC Berkeley {raaz.rsk, minhnhat, koulik}@berkeley.edu Martin J. Wainwright UC Berkeley Voleon Group wainwrig@berkeley.edu Michael I. Jordan UC Berkeley jordan@berkeley.edu
Pseudocode No The paper describes the EM algorithm mathematically but does not include structured pseudocode or an algorithm block.
Open Source Code No The paper does not provide any statement or link regarding the availability of its source code.
Open Datasets No The paper uses simulated data for its numerical examples rather than a publicly available dataset. It states, 'data is generated according to some true distribution P'.
Dataset Splits No The paper does not specify training, validation, or test dataset splits; it describes numerical examples with simulated data rather than empirical experiments on partitioned datasets.
Hardware Specification No The paper does not provide any specific details about the hardware used to run the experiments or simulations.
Software Dependencies No The paper does not specify any software dependencies with version numbers used for the numerical examples.
Experiment Setup Yes Specific parameters for the numerical examples are given, such as 'Case 1: θ = 5, ρ = 0.2', 'Case 2: θ = 5, ω = 0.2', 'Case 3: θ = 0.5', and 'starting point θ0 B(θ, θ /4)'.