A Stochastic Linearized Augmented Lagrangian Method for Decentralized Bilevel Optimization

Authors: Songtao Lu, Siliang Zeng, Xiaodong Cui, Mark Squillante, Lior Horesh, Brian Kingsbury, Jia Liu, Mingyi Hong

NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical results tested on multi-agent RL problems showcase the superiority of SLAM compared with the benchmarks. In this section, we evaluate our proposed algorithm using two MARL environments: 1) the cooperative navigation task [16], which is built on the Open AI Gym platform [46]; and 2) the pursuit-evasion game [47], which is built on the Petting Zoo platform [48].
Researcher Affiliation Collaboration IBM Research, Thomas J. Watson Research Center, Yorktown Heights, NY 10598 songtao@ibm.com,{cuix,mss,lhoresh,bedk}@us.ibm.com Dept. of Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN 55455 {zeng0176,mhong}@umn.edu Dept. of Electrical and Computer Engineering, The Ohio State University, Columbus, OH 43210 liu@ece.osu.edu
Pseudocode Yes Algorithm 1 Decentralized implementation of SLAM; Algorithm 2 Decentralized implementation of SLAM-L; Algorithm 3 Decentralized implementation of SLAM-U
Open Source Code No The paper's checklist explicitly states: 'Did you include the code, data, and instructions needed to reproduce the main experimental results (either in the supplemental material or as a URL)? [No]'
Open Datasets Yes In this section, we evaluate our proposed algorithm using two MARL environments: 1) the cooperative navigation task [16], which is built on the Open AI Gym platform [46]; and 2) the pursuit-evasion game [47], which is built on the Petting Zoo platform [48].
Dataset Splits No The paper does not specify exact train/validation/test splits, percentages, or sample counts in the main text. It defers 'Detailed experimental settings' to the supplement.
Hardware Specification No The paper does not specify any hardware details such as GPU/CPU models, memory, or specific computing clusters used for the experiments in the main text.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies or libraries used in the experiments.
Experiment Setup No The paper states: 'Detailed experimental settings and additional numerical results are provided in the supplement.' While it mentions some high-level details like the number of agents and network types, it does not provide concrete hyperparameters (e.g., learning rates, epochs, optimizer settings) in the main text.