Multi-Agent Determinantal Q-Learning

Authors: Yaodong Yang, Ying Wen, Jun Wang, Liheng Chen, Kun Shao, David Mguni, Weinan Zhang

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate our algorithm on various cooperative benchmarks; its effectiveness has been demonstrated when compared with the state-of-the-art. 4. Experiments We compare Q-DPP with state-of-the-art CTDE solvers for multi-agent cooperative tasks, including COMA (Foerster et al., 2018), VDN (Sunehag et al., 2017), QMIX (Rashid et al., 2018), QTRAN (Son et al., 2019), and MAVEN (Mahajan et al., 2019).
Researcher Affiliation Collaboration 1Huawei Technology R&D UK. 2University College London. 3Shanghai Jiaotong University.
Pseudocode Yes Algorithm 1 Multi-Agent Determinantal Q-Learning
Open Source Code Yes Code is released in https://github. com/QDPP-Git Hub/QDPP.
Open Datasets Yes We consider four cooperative tasks in Fig. 3, all of which require non-trivial value function decomposition to achieve the largest reward. All baselines are imported from Py MARL (Samvelyan et al., 2019).
Dataset Splits No The paper mentions training on 'sampled mini-batch of transition data' but does not specify any explicit train/validation/test dataset splits, percentages, or methodology for partitioning data for reproducibility.
Hardware Specification No The paper does not provide specific details regarding the hardware used for experiments, such as CPU or GPU models, or cloud computing specifications.
Software Dependencies No The paper mentions using PyMARL for baselines, but it does not provide specific version numbers for PyMARL or any other software dependencies needed to replicate the experiments.
Experiment Setup Yes Detailed settings are in Appendix B.