A Probabilistic Programming Approach To Probabilistic Data Analysis

Authors: Feras Saad, Vikash K. Mansinghka

NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental empirical illustrations of the efficacy of the framework for analyzing a real-world database of Earth satellites.Second, it reports the lines of code and accuracy of CGPMs compared with baseline solutions from standard machine learning libraries.
Researcher Affiliation Academia Feras Saad MIT Probabilistic Computing Project fsaad@mit.edu Vikash Mansinghka MIT Probabilistic Computing Project vkm@mit.edu
Pseudocode Yes Algorithm 2a simulate for CGPMs in a probabilistic programming environment.Algorithm 2b logpdf for CGPMs in a probabilistic programming environment.Algorithm 3a simulate in a directed acyclic network of CGPMs.Algorithm 3b logpdf in a directed acyclic network of CGPMs.Algorithm 3c Weighted forward sampling in a directed acyclic network of CGPMs.
Open Source Code No The paper references Bayes DB [10] and Venture Script [8], which are platforms used, but it does not provide a direct link or explicit statement for the open-sourcing of the specific code developed for the methodology in this paper.
Open Datasets Yes a database of 1163 satellites maintained by the Union of Concerned Scientists [12].U. of Concerned Scientists. UCS Satellite Database, 2015.
Dataset Splits No The paper mentions "data_train" and "data_test" but does not provide specific details on validation splits (e.g., percentages, sample counts, or methodology for generating them).
Hardware Specification No The paper does not provide any specific hardware details (e.g., GPU/CPU models, memory, or cloud computing instance types) used for running its experiments.
Software Dependencies No The paper mentions "Venture Script" and "Bayes DB" but does not provide specific version numbers for these or any other software dependencies crucial for replication.
Experiment Setup Yes INITIALIZE 10 MODELS FOR satellites_hybrid; ANALYZE satellites_hybrid FOR 100 ITERATIONS; OVERRIDE GENERATIVE MODEL FOR type_of_orbit ... USING RANDOM_FOREST (num_categories = 7); OVERRIDE GENERATIVE MODEL FOR launch_mass_kg, dry_mass_kg, power_watts, perigee_km, apogee_km USING FACTOR_ANALYSIS (dimensionality = 2);