Spectral MLE: Top-K Rank Aggregation from Pairwise Comparisons

Authors: Yuxin Chen, Changho Suh

ICML 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental The practical applicability of Spectral MLE is further corroborated by numerical experiments.
Researcher Affiliation Academia Yuxin Chen YXCHEN@STANFORD.EDU Department of Statistics, Stanford University, Stanford, CA 94305, USA Changho Suh CHSUH@KAIST.AC.KR Department of Electrical Engineering, Korea Advanced Institute of Science and Technology, Korea
Pseudocode Yes Algorithm 1 Spectral MLE.
Open Source Code No The paper does not explicitly state that the source code for its methodology is made publicly available or provide a link to a repository.
Open Datasets No The latent scores are generated uniformly over [0.5, 1]. For each (pobs, L), the paired comparisons are randomly generated as per the BTL model, and we perform score inference by means of both Rank Centrality and Spectral MLE. The generated data is not made publicly available with access information.
Dataset Splits No In all numerical simulations performed here, we pick c2 = 5 and c3 = 1, and do not split samples.
Hardware Specification No The paper describes synthetic experiments but does not provide any specific details about the hardware used to conduct them.
Software Dependencies No The paper describes algorithms and numerical experiments but does not specify any software libraries or their version numbers used in the implementation.
Experiment Setup Yes In all numerical simulations performed here, we pick c2 = 5 and c3 = 1, and do not split samples. We focus on the case where n = 100, where each reported result is calculated by averaging over 200 Monte Carlo trials. The latent scores are generated uniformly over [0.5, 1]. For each (pobs, L), the paired comparisons are randomly generated as per the BTL model.