Monotone Retargeted Preference Completion
Contact: Suriya Gunasekar (firstname.lastname@example.org)
This is a reference implementation of the algorithm described in the paper:
Suriya Gunasekar, Oluwasanmi Koyejo, and Joydeep Ghosh, "Preference Completion from Partial Rankings" NIPS 2016.
It is being made available without any guarantees for non-commercial research use only. If you find this code useful in your research, please consider citing the paper.
smc.m: function implementing standard matrix completion rmc_fixed_margin.m: function implementing the proposed monorone retargetted preference completion algorithm.
movienlens.m and neurosynth.m scripts to run smc and rmc on movielens and neurosynth datasets (Warning: has some hardcoded numbers and paths)
Movielens/: Folder with scripts to process data and results of Movielens dataset Neurosynth/: Folder with scripts to process data and results of Neurosynth dataset