The Limits of Maxing, Ranking, and Preference Learning

Authors: Moein Falahatgar, Ayush Jain, Alon Orlitsky, Venkatadheeraj Pichapati, Vaishakh Ravindrakumar

ICML 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We present experiments over simulated data in Section 8 and end with our conclusions in Section 9.
Researcher Affiliation Academia 1University of California, San Diego. Correspondence to: Venkatadheeraj Pichapati <dheerajpv7@ucsd.edu>.
Pseudocode Yes Algorithm 1 SOFT-SEQ-ELIM; Algorithm 2 NEAR-OPT-MAX; Algorithm 3 OPT-MAX-LOW; Algorithm 4 OPT-MAX; Algorithm 5 APPROX-PROB
Open Source Code No The paper does not contain any explicit statement about releasing source code for the methodology described, nor does it provide a link to a code repository.
Open Datasets No The experiments are conducted using 'simulated data' as stated in Section 8: 'We present experiments over simulated data in Section 8 and end with our conclusions in Section 9.' There is no mention of a publicly available dataset with concrete access information like a link, DOI, or formal citation.
Dataset Splits No The paper describes experiments over simulated data but does not provide specific training/validation/test dataset splits, sample counts, or references to predefined splits.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, processor types, or memory amounts used for running the experiments. It only states that experiments were run on 'simulated data'.
Software Dependencies No The paper does not provide any specific ancillary software details, such as library or solver names with version numbers, needed to replicate the experiments.
Experiment Setup Yes In all experiments, we use maxing algorithms to find a 0.05-maximum with δ = 0.1. All results presented here are averaged over 1000 runs.