Polynomial-Delay MAG Listing with Novel Locally Complete Orientation Rules

Authors: Tian-Zuo Wang, Wen-Bo Du, Zhi-Hua Zhou

ICML 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we evaluate the effectiveness and efficiency of the proposed MAG listing algorithm. Our method, MAGLIST-POLY, is compared with MAGLIST and BRUTEFORCE, which are presented by Wang et al. (2024a). ... The experimental results are shown in Fig. 9. When d < 10, all the methods return identical set of MAGs, demonstrating the effectiveness of MAGLIST-POLY. Moreover, MAGLIST-POLY consistently outperforms the other methods in runtime, highlighting its efficiency.
Researcher Affiliation Academia 1National Key Laboratory for Novel Software Technology, Nanjing University, China 2School of Artificial Intelligence, Nanjing University, China. Correspondence to: Zhi-Hua Zhou <EMAIL>.
Pseudocode Yes Algorithm 1 MAGLIST-POLY Require: A PAG P 1: S = Record all the MAGs consistent with P 2: ORIENTGRAPH(P) 3: function ORIENTGRAPH(M, S) 4: if there are no circles in M then M is a MAG 5: S S {M} 6: else 7: Select a variable X where there are circles in M 8: I = Record all the PMGs without circles at X obtained from M 9: LOCALTRANSFORM(M, X) 10: for M in I do 11: ORIENTGRAPH(M , S) 12: end for 13: end if 14: end function 15: function LOCALTRANSFORM(H, X) 16: if there are no circles at X in H then 17: I I {H} 18: else 19: Select a vertex V with an edge X V in H 20: Obtain H from H by orienting X V and using the proposed orientation rules 21: LOCALTRANSFORM(H , X) 22: Obtain H from H by orienting X V and using the proposed orientation rules 23: LOCALTRANSFORM(H , X) 24: end if 25: end function Ensure: S
Open Source Code No The text does not contain a concrete statement about open-sourcing the code or a link to a code repository for the methodology described.
Open Datasets Yes Finally, we conduct an experiment using real data processed from Wang et al. (2017) consisting of 7466 measurements of the abundance of phosphoproteins and phospholipids recorded under various experimental conditions.
Dataset Splits No The paper describes generating synthetic graphs and using real data from a previous study (Wang et al., 2017), but does not provide specific training/test/validation dataset split information.
Hardware Specification No The paper does not provide specific hardware details used for running its experiments.
Software Dependencies No The paper does not provide specific ancillary software details with version numbers needed to replicate the experiment.
Experiment Setup No The paper describes parameters for synthetic data generation (number of vertices d and graph density ρ) and the procedure for real data experiments (2 interventions, repeated 10 times), but does not provide specific hyperparameters or system-level training settings for a machine learning model.