FoX: Formation-Aware Exploration in Multi-Agent Reinforcement Learning

Authors: Yonghyeon Jo, Sunwoo Lee, Junghyuk Yeom, Seungyul Han

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Numerical results show that the proposed Fo X framework significantly outperforms the state-of-the-art MARL algorithms on Google Research Football (GRF) and sparse Starcraft II multi-agent challenge (SMAC) tasks.
Researcher Affiliation Academia Artificial Intelligence Graduate School, UNIST, Ulsan, South Korea
Pseudocode Yes We summarize the proposed Fo X framework as Algorithm 1 and Figure 5.
Open Source Code Yes The source code of our proposed algorithm is available at https://github.com/hyeon1996/Fo X.
Open Datasets Yes In our experiments, we evaluate the performance of the proposed Fo X framework in the challenging cooperative multi-agent environments: the sparse Star Craft multi-agent challenge (SMAC) (Samvelyan et al. 2019) and Google Research Football (GRF) (Kurach et al. 2020).
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning for validation.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types with speeds, memory amounts, or detailed computer specifications) used for running its experiments.
Software Dependencies No The paper mentions various MARL algorithms and environments (e.g., QMIX, SMAC, GRF) used in the experiments, but does not provide specific ancillary software details like library or solver names with version numbers.
Experiment Setup Yes For hyperparameters β1, β2, we have tested β1 {0.001, 0.005, 0.01, 0.02, 0.1} and β2 {0.001, 0.005, 0.01, 0.05} for both environments.