Information Theoretical Estimators Toolbox

Authors: Zoltán Szabó

JMLR 2014 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present ITE (information theoretical estimators) a free and open source, multi-platform, Matlab/Octave toolbox that is capable of estimating many different variants of entropy, mutual information, divergence, association measures, cross quantities, and kernels on distributions... ITE also includes a prototype application in a central problem class of signal processing, independent subspace analysis and its extensions. ... Quick Tests: Beyond IPA, ITE provides quick tests to study the efficiency of the estimators. These tests cover (i) analytical value vs. estimation, (ii) positive semi-definiteness of Gram matrices defined by distribution kernels and (iii) image registration problems.
Researcher Affiliation Academia Zoltán Szabó EMAIL Gatsby Computational Neuroscience Unit Centre for Computational Statistics and Machine Learning University College London Alexandra House, 17 Queen Square, London WC1N 3AR
Pseudocode Yes >Y1 = rand(3,1000); Y2 = rand(3,2000); %data of interest >mult = 1; %multiplicative constant is important >co = D_initialization( Jdistance ,mult);%initialize the estimator >D = D_estimation(Y1,Y2,co); %estimation Next, we demonstrate how one can construct meta estimators in ITE. We consider the definitions of the initialization and the estimation of the J-distance. The KLdivergence, which is symmetrised in J-distance, is estimated based on the existing k-nearest neighbor technique. function [co] = DJdistance_initialization(mult) co.name = Jdistance ; %name of the estimator co.mult = mult; %importance of multiplicative const. co.member_name = KL_k NN_k ; %method used for KL estimation co.member_co = D_initialization(co.member_name,mult); %initialization function [D_J] = DJdistance_estimation(X,Y,co) D_J = D_estimation(X,Y,co.member_co) + D_estimation(Y,X,co.member_co);
Open Source Code Yes We present ITE (information theoretical estimators) a free and open source, multi-platform, Matlab/Octave toolbox... ITE is released under the free and open source GNU GPLv3 (...) license. ... The ITE package is available at https://bitbucket.org/szzoli/ite/.
Open Datasets No The paper describes a software toolbox and its capabilities. While it mentions quick tests and a prototype application, it does not specify any publicly available datasets used for these, nor does it provide links, DOIs, or formal citations for any data.
Dataset Splits No The paper does not describe the use of any specific datasets, other than randomly generated data for demonstration purposes, and therefore does not provide any dataset split information.
Hardware Specification No The paper mentions that 'a single general desktop computer can handle about a million observations' but does not provide any specific details about the hardware, such as CPU, GPU models, or memory specifications.
Software Dependencies No The paper states that the ITE package 'only needs a Matlab or an Octave environment with standard toolboxes' and suggests 'Visual C++ (GCC) compiler' on Windows (Linux), but it does not provide specific version numbers for any of these software dependencies.
Experiment Setup No The paper provides code examples to demonstrate the use of the toolbox but does not specify any concrete hyperparameter values, training configurations, or system-level settings for experimental reproduction.