Pairwise Choice Markov Chains

Authors: Stephen Ragain, Johan Ugander

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

Reproducibility Variable Result LLM Response
Research Type Experimental We show that the PCMC model significantly outperforms both the Multinomial Logit (MNL) model and a mixed MNL (MMNL) model in prediction tasks on both synthetic and empirical datasets known to exhibit violations of Luce s axiom.
Researcher Affiliation Academia Stephen Ragain Management Science & Engineering Stanford University Stanford, CA 94305 sragain@stanford.edu Johan Ugander Management Science & Engineering Stanford University Stanford, CA 94305 jugander@stanford.edu
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code Yes Data and code available here: https://github.com/sragain/pcmc-nips
Open Datasets Yes We evaluate our inference procedure on two empirical choice datasets, SFwork and SFshop, collected from a survey of transportation preferences around the San Francisco Bay Area [16].
Dataset Splits Yes The results are averaged over 1,000 different permutations of the data with a 75/25 train/test split employed for each permutation.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory) used for running experiments.
Software Dependencies No The paper mentions using 'SLSQP [25]' for optimization, but it does not provide specific version numbers for this or any other software dependencies.
Experiment Setup Yes We applied small amounts of additive smoothing to each dataset. We used a discrete mixture of k MNL models (with O(kn) parameters), choosing k so that the MMNL model had strictly more parameters than the PCMC model on each data set.