Near-Optimal Smoothing of Structured Conditional Probability Matrices

Authors: Moein Falahatgar, Mesrob I. Ohannessian, Alon Orlitsky

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

Reproducibility Variable Result LLM Response
Research Type Experimental 7 Experiments Having expounded the theoretical merit of properly smoothing structered conditional probability matrices, we give a brief empirical study of its practical impact. We use both synthetic and real data.
Researcher Affiliation Academia Moein Falahatgar University of California, San Diego San Diego, CA, USA moein@ucsd.edu Mesrob I. Ohannessian Toyota Technological Institute at Chicago Chicago, IL, USA mesrob@ttic.edu Alon Orlitsky University of California, San Diego San Diego, CA, USA alon@ucsd.edu
Pseudocode Yes Algorithm: ADD1 2-SMOOTHED LOW-RANK
Open Source Code No The paper does not provide any concrete access to source code (e.g., a specific repository link, an explicit code release statement, or code in supplementary materials) for the methodology described.
Open Datasets Yes tartuffe, a French text, train and test size: 9.3k words, vocabulary size: 2.8k words. genesis, English version, train and test size: 19k words, vocabulary size: 4.4k words brown, shortened Brown corpus, train and test size: 20k words, vocabulary size: 10.5k words All but the first one are readily available through the Python NLTK
Dataset Splits Yes In particular, half of the data was held out as a validation set, and for a range of different choices for m, the model was trained and its cross-entropy on the validation set was calculated.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, processor types, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details (e.g., library or solver names with version numbers) needed to replicate the experiment.
Experiment Setup Yes For all these experiments, m = 50 and 200 iterations were performed.