Inductive Pairwise Ranking: Going Beyond thenlog(n) Barrier

Authors: U.N. Niranjan, Arun Rajkumar

AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In practice, through systematic synthetic simulations, we confirm our theoretical findings regarding improvements in the sample complexity due to the use of feature information. Moreover, on popular real-world preference learning datasets, with as less as 10% sampling of the pairwise comparisons, our method recovers a good ranking.
Researcher Affiliation Collaboration U. N. Niranjan Department of Computer Science University of California Irvine un.niranjan@uci.edu. Arun Rajkumar Data Analytics Lab Xerox Research Centre India arun.rajkumar@xerox.com. Part of the work done while interning at Xerox Research Centre India.
Pseudocode Yes Algorithm 3 IPR: Inductive Pairwise Ranking. Subroutine 1 IMC: Inductive Matrix Completion. Subroutine 2 PR: Pairwise Ranking (Copeland Procedure).
Open Source Code No The paper does not provide any explicit statements about releasing source code or links to a code repository.
Open Datasets Yes 1. Sushi: This data (Kamishima and Akaho 2009) is from a survey of 5000 customers. [...] 2. Car: The task in this dataset (Abbasnejad et al. 2013) is find an order of preference among ten cars.
Dataset Splits No The paper mentions "10% sampling" in the abstract and discussion of varying 'm' (number of pairs compared) for synthetic data, but does not provide explicit train/validation/test splits, percentages, or absolute counts for reproducibility of data partitioning.
Hardware Specification No The paper does not specify any hardware details such as GPU models, CPU types, or memory used for running the experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes For IPR, we choose λL = 10-2 and λN = 102.