Instance-specific algorithm configuration via unsupervised deep graph clustering

• Main Authors: SONG, Wen, LIU, Yi, CAO, Zhiguang, WU, Yaoxin, LI, Qiqiang
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2023
• Subjects: Algorithm configuration; Unsupervised graph embedding; Mixed-integer programming; Artificial Intelligence and Robotics; Theory and Algorithms
• Online Access: https://ink.library.smu.edu.sg/sis_research/8086; https://ink.library.smu.edu.sg/context/sis_research/article/9089/viewcontent/1_s2.0_S0952197623009247_pvoa_cc_by.pdf
• Institution: Singapore Management University

Instance-specific Algorithm Configuration (AC) methods are effective in automatically generating high-quality algorithm parameters for heterogeneous NP-hard problems from multiple sources. However, existing works rely on manually designed features to describe training instances, which are simple num...