Multi-Agent Actor-Critic with Hierarchical Graph Attention Network

Authors: Heechang Ryu, Hayong Shin, Jinkyoo Park7236-7243

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

Reproducibility Variable Result LLM Response
Research Type Experimental Empirically, we demonstrate that the proposed model outperforms existing methods in several mixed cooperative and competitive tasks.
Researcher Affiliation Academia Heechang Ryu, Hayong Shin, Jinkyoo Park Industrial & Systems Engineering, KAIST, Republic of Korea {rhc93, hyshin, jinkyoo.park}@kaist.ac.kr
Pseudocode No The paper describes the methods using prose and mathematical formulas but does not include any structured pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any statements about open-source code availability or links to a code repository.
Open Datasets No The paper describes custom game environments (Cooperative Navigation, Predator-Prey games) without providing access information (links, DOIs, or formal citations with author/year for public datasets) for the data or environments used.
Dataset Splits No The paper mentions training and testing phases (e.g., "in training, as shown in Figure 3", "during testing with 200 episodes") but does not explicitly provide specific train/validation/test dataset split percentages or sample counts.
Hardware Specification No The paper does not specify any hardware details (e.g., GPU/CPU models, memory) used for running the experiments.
Software Dependencies No The paper does not specify any software dependencies with version numbers (e.g., programming languages, libraries, frameworks, or solvers).
Experiment Setup No The paper describes the network architecture (two-layered MLP with 256 units and Re LUs for embedding and attention functions) and total training steps (3 million steps) but does not provide specific hyperparameters such as learning rate, batch size, or optimizer settings.