Notice: This website is continuously updated as new conference proceedings and journal volumes are published. Consequently, the figures presented here may differ from those reported in our publications.
A brief overview of our methodology is provided on the Methods page, along with references to publications documenting our approach and links to the code and data used in each study.
The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates.
Value-Based Deep Multi-Agent Reinforcement Learning with Dynamic Sparse Training
Authors: Pihe Hu, Shaolong Li, Zhuoran Li, Ling Pan, Longbo Huang
NeurIPS 2024 | Venue PDF | LLM Run Details | Input Tokens: 39,457 Total number of tokens sent to the LLM as input for this paper's analysis. | Output Tokens: 5,261 Total number of tokens produced by the LLM (including reasoning/thinking tokens) for this paper's analysis.
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Our comprehensive experimental investigation across various value-based MARL algorithms on multiple benchmarks demonstrates, for the first time, significant reductions in redundancy of up to 20 in Floating Point Operations (FLOPs) for both training and inference, with less than 3% performance degradation. |
| Researcher Affiliation | Academia | Pihe Hu Tsinghua University Beijing, China EMAIL Shaolong Li Central South University Changsha, China EMAIL Zhuoran Li Tsinghua University Beijing, China EMAIL Ling Pan Hong Kong University of Science and Technology Hong Kong, China EMAIL Longbo Huang Tsinghua University Beijing, China EMAIL |
| Pseudocode | Yes | Algorithm 1 Topology Evoltion[Evci et al., 2020]; Algorithm 2 MAST-QMIX; Algorithm 3 MAST-(OW)QMIX |
| Open Source Code | No | The code will be open-sourced upon publication of the paper. |
| Open Datasets | Yes | In this section, we conduct a comprehensive performance evaluation of MAST across various tasks in the Star Craft Multi-Agent Challenge (SMAC) [Samvelyan et al., 2019] benchmark. Additional experiments on the multi-agent Mu Jo Co (MAMu Jo Co) [Peng et al., 2021] benchmark are provided in Appendix B.9. |
| Dataset Splits | Yes | In this section, we conduct a comprehensive performance evaluation of MAST across various tasks in the Star Craft Multi-Agent Challenge (SMAC) [Samvelyan et al., 2019] benchmark. [...] The environments are tested using their default configurations, with other settings following FACMAC [Peng et al., 2021]. |
| Hardware Specification | Yes | Our experiments are implemented with Py Torch 2.0.0 [Paszke et al., 2017] and run on 4 NVIDIA GTX Titan X (Pascal) GPUs. |
| Software Dependencies | Yes | Our experiments are implemented with Py Torch 2.0.0 [Paszke et al., 2017] |
| Experiment Setup | Yes | Table 3 provides a comprehensive overview of the hyperparameters employed in our experiments for MAST-QMIX, MAST-WQMIX, and MAST-RES. It includes detailed specifications for network parameters, RL parameters, and topology evolution parameters, allowing for a thorough understanding of our configurations. |