A Stochastic Path Integral Differential EstimatoR Expectation Maximization Algorithm

Authors: Gersende Fort, Eric Moulines, Hoi-To Wai

NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical results support our findings.
Researcher Affiliation Academia Gersende Fort Institut de Math ematiques de Toulouse Universit e de Toulouse; CNRS UPS, Toulouse, France gersende.fort@math.univ-toulouse.fr Eric Moulines Centre de Math ematiques Appliqu ees Ecole Polytechnique, France CS Dpt, HSE University, Russian Federation eric.moulines@polytechnique.edu Hoi-To Wai Department of SEEM The Chinese University of Hong Kong Shatin, Hong Kong htwai@cuhk.edu.hk
Pseudocode Yes Algorithm 1: The SPIDER-EM algorithm.
Open Source Code No The paper does not provide an explicit statement or link for open-source code availability.
Open Datasets Yes MNIST Dataset. We perform experiment on the MNIST dataset to illustrate the effectiveness of SPIDER-EM on real data; this example is taken from [23, Section 5].
Dataset Splits No The paper uses the MNIST dataset but does not specify training, validation, or test splits. It mentions preprocessing steps and estimating a GMM model.
Hardware Specification No The paper does not provide any specific hardware details used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes For SPIDER-EM, we set b = dpn/20e, kin = dn/be and a fixed step size γk = 0.01. The minibatch size is set to be b = 100 and the stepsize γ = 5 10 3 except for i EM where γ = 1. The same initial value b Sinit is used for all experiments. We have implemented the procedure of [19] in order to obtain the initialization init and then we set b Sinit def = s( init) ( W(b Sinit) = 58.3).