Preference Completion from Partial Rankings

Authors: Suriya Gunasekar, Oluwasanmi O. Koyejo, Joydeep Ghosh

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

Reproducibility Variable Result LLM Response
Research Type Experimental We further show promising empirical results for a novel and challenging application of collaboratively ranking of the associations between brain regions and cognitive neuroscience terms.
Researcher Affiliation Academia Suriya Gunasekar University of Texas, Austin, TX, USA suriya@utexas.edu Oluwasanmi Koyejo University of Illinois, Urbana-Champaign, IL, USA sanmi@illinois.edu Joydeep Ghosh University of Texas,Austin, TX, USA ghosh@ece.utexas.edu
Pseudocode Yes Algorithm 1 Proximal Gradient Descent for (2) with input Ω, {y(j) j }, ϵ and paramter λ
Open Source Code No The paper does not provide a direct link or explicit statement about the availability of the source code for the methodology described in this paper. It mentions
Open Datasets Yes Movielens is a movie recommendation website administered by Group Lens Research. We used competitive benchmarked movielens 100K dataset. [...] Neurosynth[37] is a publicly available database consisting of data automatically extracted from a large collection of functional magnetic resonance imaging (f MRI) publications...
Dataset Splits Yes We used the 5 fold train/test splits provided with the dataset (the test splits are non-overlapping). [...] We used 10% of randomly sampled entries of the matrix as test data and a another 10% for validation. We created training datasets at various sample sizes by subsampling from the remaining 80% of the data. This random split is replicated multiple times to obtain 3 bootstrapped datasplits (note that unlike cross validation, the test datasets here can have some overlapping entries).
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, memory amounts, or cloud instance types used for running its experiments.
Software Dependencies No The paper mentions "nilearn python package" in the context of data preprocessing for the Neurosynth dataset, but it does not specify a version number for nilearn or any other software dependency.
Experiment Setup No The paper mentions that "hyperparameters were tuned using grid search on a logarithmic scale" for SMC and MRPC, but it does not provide the specific values or ranges of these hyperparameters or other detailed training configurations. It states that for CoFiRank, default parameters were used due to computational cost, but this is a baseline, not their method.