VBPR: Visual Bayesian Personalized Ranking from Implicit Feedback

Authors: Ruining He, Julian McAuley

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

Reproducibility Variable Result LLM Response
Research Type Experimental Experiments In this section, we perform experiments on multiple realworld datasets.
Researcher Affiliation Academia Ruining He University of California, San Diego r4he@ucsd.edu Julian Mc Auley University of California, San Diego jmcauley@ucsd.edu
Pseudocode No The paper describes the model learning process using mathematical equations and textual explanations, but no structured pseudocode or clearly labeled algorithm block is present.
Open Source Code No All of our code and datasets shall be made available at publication time so that our experimental evaluation is completely reproducible.
Open Datasets Yes The first group of datasets are from Amazon.com introduced by Mc Auley et al. (2015). ... We also introduce a new dataset from Tradesy.com... all of which shall be made available at publication time.
Dataset Splits Yes We split our data into training/validation/test sets by selecting for each user u a random item to be used for validation Vu and another for testing Tu. All remaining data is used for training.
Hardware Specification No All experiments were performed on a standard desktop machine with 4 physical cores and 32GB main memory.
Software Dependencies No The paper mentions software components like the "Caffe reference model" and "My Media Lite" but does not provide specific version numbers for these or any other ancillary software components used in the experiments.
Experiment Setup Yes All hyperparameters are tuned using a validation set as we describe in our experimental section later. On Amazon, regularization hyperparamter λΘ = 10 works the best for BPR-MF, MM-MF and VBPR in most cases. While on Tradesy.com, λΘ = 0.1 is set for BPR-MF and VBPR and λΘ = 1 for MM-MF. λE is always set to 0 for VBPR. For IBR, the rank of the Mahalanobis transform is set to 100, which is reported to perform very well on Amazon data.