Dueling Bandits with Qualitative Feedback

Authors: Liyuan Xu, Junya Honda, Masashi Sugiyama5549-5556

AAAI 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We test the empirical performance of the proposed algorithms through experiments based on both synthetic setting and real-world data.
Researcher Affiliation Academia Liyuan Xu,1,2 Junya Honda,1,2 Masashi Sugiyama1,2 1The University of Tokyo, 2RIKEN liyuan@ms.k.u-tokyo.ac.jp, honda@stat.t.u-tokyo.ac.jp, sugi@k.k.u-tokyo.ac.jp
Pseudocode Yes Algorithm 1: Thompson Condorcet sampling
Open Source Code No The paper mentions '1The longer version including all appendices is available at https://arxiv.org/abs/1809.05274' which is a link to an arXiv preprint, but does not explicitly state that the source code for the described methodology is available at this link or elsewhere.
Open Datasets Yes We used two web search datasets. The first is the MSLRWEB10K dataset (Qin et al. 2010), which consists of 10,000 search queries over the documents from search results. ... The other is the MQ2008 dataset (Qin and Liu 2013)...
Dataset Splits No The paper mentions using 'MSLR-WEB10K dataset' and 'MQ2008 dataset' and that it 'repeat[s] 100 runs for each instance', but it does not provide specific details on train/validation/test splits (percentages, sample counts, or explicit references to predefined splits).
Hardware Specification No No specific hardware details (e.g., exact GPU/CPU models, processor types, memory amounts, or detailed computer specifications) used for running experiments are provided in the paper.
Software Dependencies No The paper does not provide specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiment.
Experiment Setup Yes We set t0 = 10, and the Figure 1 is the experimental result when the number of rankers is K = 5.