Approximating weighted completion time via stronger negative correlation
Author(s)
Baveja, Alok; Qu, Xiaoran; Srinivasan, Aravind
Download10951_2023_780_ReferencePDF.pdf (485.6Kb)
Publisher Policy
Publisher Policy
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Terms of use
Metadata
Show full item recordAbstract
Minimizing the weighted completion time of jobs in the unrelated parallel machines model is a fundamental scheduling problem. The first (3/2−𝑐)
–approximation algorithm for this problem, for some constant 𝑐>0
, was obtained in the work of Bansal et al. (SIAM J Comput, 2021). A key ingredient in this work was the first dependent-rounding algorithm with a certain guaranteed amount of negative correlation. We improve upon this guaranteed amount from 1/108 to 1/27, thus also improving upon the constant c in the algorithms of Bansal et al. and Li (SIAM J Comput, 2020) for weighted completion time. Given the now-ubiquitous role played by dependent rounding in scheduling and combinatorial optimization, our improved dependent rounding is also of independent interest.
Date issued
2023-03-30Journal
Journal of Scheduling
Publisher
Springer US
Citation
Baveja, A., Qu, X. & Srinivasan, A. Approximating weighted completion time via stronger negative correlation. J Sched 27, 319–328 (2024).
Version: Author's final manuscript